MASTERUPPSATS I KOGNITIONSVETENSKAP

Barriers to Near-miss Reporting in the Maritime Domain

Fredrik Köhler

2010-12-08

Institutionen för datavetenskap

Linköpings universitet

Handledare: Margareta Lützhöft och Gesa Praetorius, Chalmers tekniska

högskola

ISRN: LIU-IDA/KOGVET-A--10/014--SE

Abstract The catastrophic accident of the ferry Herald of Free Enterprise made it clear that the development

of accident prevention in the maritime domain must not only rely on negative events but rather on

proactive measures.

Near-miss reporting is becoming widespread as a proactive tool for accident prevention in various

domains. This thesis aims to examine and identify barriers to near-miss reporting through studying

the national reporting system INSJÖ and local company specific systems in the Swedish maritime

domain.

Interviews with representatives from Swedish shipping companies (designated persons responsible

for safety work in each company and officers responsible for the reporting at sea) were conducted as

a means of data collection. Based on the data two separate analyses were made; one in a naturalistic

fashion and one using a framework of barriers and incentives derived from various social technical

domains in which near-miss reporting has been institutionalized.

The results of the two analyses highlight differences regarding how and with whom information

should be shared. The therapeutic factor, to teach and learn from others was emphasized as

important by the majority of the interviewees. Further, potential external influences, issues

concerning anonymity and the risk of rehearsed benefits of reporting are also made visible. Finally,

critique against the accident-ratio models, that introduced the near-miss concept, is presented and it

is argued that these models might be too simplistic to explain why accidents occur.

It is concluded that, in order to create effective reporting systems and to decrease the risk of creating

a disparity between rehearsed benefits and how the system is used in reality, it is important to give

the personnel ownership of their own reporting system and the knowledge of how and why to use it.

Nevertheless, near-miss reporting might be used as a powerful tool and incentive for proactive work

and accident prevention.

Acknowledgement Firstly, I would like to thank my supervisors Margareta Lützhöft and Gesa Praetorius from Chalmers

University of Technology. They have both been excellent supervisors on the long and occasionally

rocky road to completion of this thesis. Thank you for your dedication, help and encouragement.

Further, I would like to thank Olle Bråfelt at ICC, IPSO Classification & Control AB for appreciated and

useful support at the beginning of this thesis.

Of course, I would also like to thank Natalia González for an excellent opposition of this thesis.

I would also like to thank David Wetterbro for reading and commenting on several drafts of this

thesis and Caroline Lindström for all your support and cheering words.

Finally, I would like to thank all the DPs and master mariners participating in this study, without you

this thesis would not have been possible.

Table of Contents 1 Introduction ...................................................................................................................................1

1.1 Outline of this thesis ...................................................................................................................... 1

2 Background ....................................................................................................................................3

2.1 The Maritime Domain ................................................................................................................... 3

2.2 Definitions of accident, incident and near-miss ............................................................................ 4

2.3 Reporting systems ......................................................................................................................... 8

3 Aim.............................................................................................................................................. 17

3.1 Research question ....................................................................................................................... 17

3.2 Boundaries of the study .............................................................................................................. 17

3.3 Scope of this thesis ...................................................................................................................... 17

4 Method ....................................................................................................................................... 18

4.1 Interviews as method .................................................................................................................. 18

4.2 Data collection ............................................................................................................................. 19

4.3 Data Analysis ............................................................................................................................... 20

4.4 Validity, reliability and objectivity ............................................................................................... 23

5 Analysis ....................................................................................................................................... 26

5.1 Analysis 1 ..................................................................................................................................... 26

5.2 Reporting officers onboard ......................................................................................................... 26

5.3 DP personnel ............................................................................................................................... 31

5.4 Discussion - analysis 1 ............................................................................................................. 39

5.5 Analysis 2 – Barriers and incentives to reporting ........................................................................ 43

5.6 Discussion - analysis 2 ................................................................................................................. 50

6 Discussion ................................................................................................................................... 52

6.1 General discussion ....................................................................................................................... 52

6.2 Theoretical Framework ............................................................................................................... 53

6.3 Methodology ............................................................................................................................... 55

6.4 Further research .......................................................................................................................... 58

7 Conclusions ................................................................................................................................. 60

References ..................................................................................................................................... 61

Appendix A - Interview template ................................................................................................. 64

Abbreviations/Definitions

CHIRP The British Confidential Reporting Programme for Aviation and

Maritime

Classification Society A non-governmental organization that establishes and maintains

standards for the construction and classification of ships and offshore

structures

DP The designated person. Each shipping company must designate a

person in office with access to the highest level of management,

responsible of monitoring the safety and pollution aspects of the

operation of each ship

EMSA The European Maritime Safety Agency, the maritime safety agency of the European Union IMO The International Maritime Organization. IMO is the agency established

by United Nations with the task to develop and maintain a comprehensive regulatory framework for shipping worldwide

INSJÖ A Swedish database for reporting accidents, incidents, near-misses and

non-conformities ISM The International Safety Management Code MARS The Mariners' Alerting and Reporting Scheme Port State Control The inspection of foreign ships in national ports to verify that the ship

comply with international requirements concerning equipment and operation.

SAFIR An ISM reporting tool for accidents, near accidents and non-

conformities SMA The Swedish Maritime Administration SMS A Safety Management System SOLAS The International Convention for the Safety of Life at Sea SOS The reporting system SjöOlycksSystemet, a Swedish database for

accident reporting SSA The Swedish Shipowners’ Association STA The Swedish Transport Agency

1

1 Introduction On the 6th of March 1987 the passenger and freight ferry Herald of Free Enterprise sailed from

Zeebrugge in Belgium to Dover. The Herald was manned by a crew of 80 and held numerous vehicles

and approximately 459 passengers. The voyage to Dover was expected to be made without any

complications in the prevailing good weather with light easterly breeze and very little sea or swell. At

18.24 the ship passed the outer mole and increased its speed with the result that water came over

the bow sill and flooded into the lower car deck. Approximately four minutes later she capsized with

the dread result of 150 passengers and 38 crew members dead and many injured (MAIB, 1987). The

conclusions of the formal investigation were exceptional at the time, because not only were active

errors identified as cause to the horrific accident, but the company management’s failure of giving

proper and clear directions was also ascribed as a contributory cause to the catastrophe (Reason,

1990).

The maritime domain is one of the oldest domains that could be regarded as a socio technical

system, where technology and people interact and are dependent of each other. It precedes such

domains as aviation, the railway domain or chemical and medical industries (van der Schaaf & Kanse,

2004). Common for these domains are the importance of safety regulations concerning risk

management and having to deal with and prevent unexpected events that result in negative loss

(Barach & Small, 2000). These kinds of unexpected events can lead to dire consequences in form of

accidents and incidents that might affect both human lives and valuable property (Jones, Kirchsteiger

& Bjerke, 1999).

The formal investigation of the catastrophic accident befallen the Herald of Free Enterprise was one

of the leading reasons for new guidelines on safe operation of ships and pollution prevention by the

16th assembly of the International Maritime Organization (IMO) in 1989. The purpose of these

guidelines was to give those responsible for the operation of ships a framework for appropriate

development, implementation and assessment of safety and pollution prevention management

(IMO, 2002). These guidelines transformed into the creation of the International Management Code

for the Safe Operation of Ships and for Pollution Prevention (the ISM Code) that was adopted by the

IMO (2002) in 1993.

The Herald of Free Enterprise accident made it painfully clear that lessons learned and the

development of accident prevention must not solely rely on these kinds of utterly negative events.

Accident prevention should rather be a continuous and by large proactive process; a process which in

turn must depend on a large quantity of analysis material (Barach & Small, 2000). As means to gather

enough quantities of data, that benefit proactive accident prevention, near-miss reporting systems

are being used in several high-risk domains (Barach & Small, 2000).

1.1 Outline of this thesis This study highlights near-miss reporting and potential barriers to this kind of reporting in the

maritime domain. Chapter two, the background chapter, is divided into four sections. The first

section, 2.1, will include a short overview of the domain at hand and relevant organizations therein.

Section two, 2.2, will present the key terms accident, incident and near-miss and how they are

2

defined in previous literature. Section three, 2.3, will include a literature overview of reporting

systems and near-miss reporting in various socio technical systems. Chapter three presents the

study's aim and research question as well as the boundaries, in 3.2, and scope of the study, in 3.3.

Chapter four describes the methodological approach in section 4.1. This is followed by a presentation

of the data collection in 4.2. The section concerning the data analysis, 4.3, is divided into two

separate parts, one for each analysis. The chapter’s last section, 4.4 presents how validity, reliability

and objectivity can be applied in qualitative research. The following analysis chapter, chapter five,

presents and discusses the material found in the two analyses. Section 5.1 to 5.4 concern the first

analysis and 5.5 to 5.6 concern the second analysis. Chapter six, the discussion chapter, is divided

into four parts, a general discussion in 6.1, a discussion regarding the theoretical framework in 6.2, a

discussion regarding methodology in 6.3, and the chapter will end with suggestions of further

research in 6.4. Lastly, conclusions made are presented in chapter seven.

3

2 Background This chapter introduces the background of this study. It is divided into four sections. The first section

presents Swedish and international organizations within the maritime domain. The second

introduces the key terms accident, incident and near-miss relevant for this study. The third section

contains a cross-domain literature overview presenting barriers (and incentives) to reporting systems

in general and near-miss reporting in specific.

2.1 The Maritime Domain The focus on maritime safety, as defined by the IMO (2002), is visible on an international and

national level. This section covers organizations that are deemed important in relation to reporting

systems in the maritime domain - as a whole and to the Swedish domain in specific.

2.1.1 IMO and the role of the designated person

The International Maritime Organization was established in Geneva 1948. The main task of the

organization is the maintenance of an international regulatory framework for safety, environmental

concerns, legal matters, technical co-operation, maritime security and the efficiency of shipping

(IMO, 2002).

As mentioned previously, the International Safety Management Code (ISM) code was adopted by the

IMO in 1993, its purpose is to provide an international standard for the safe management and

operation of ships as well as for pollution prevention. The main objectives of the ISM code are to

ensure safety at sea, to prevent human injury or loss of life, and to avoid damage to the environment

(both damage to the marine environment and to property) (ISM, 2002).

The ISM code (2002) furthermore states that every shipping company should designate a person or

persons ashore that have direct access to the highest level of management. The responsibility and

authority of the designated person or persons (DP) should include monitoring all aspects concerning

the safety and pollution of the operation of each ship. The shipping company is responsible to ensure

that adequate resources and shore-based support are provided to enable the DP to carry out his or

her functions, e.g. assessing the effectiveness of Safety Management Systems (SMS), conducting

audits and overseeing regulations (IMO, 2007).

2.1.2 EMSA

The European Maritime Safety Agency (EMSA) was created in the aftermath of the Erika disaster that

occurred in 1999 - resulting in the release of thousands of tons of oil spill into the sea, polluting the

shores of Brittany, France (IMO, 2002). EMSA aims to contribute to the enhancement of the overall

maritime safety system in the European community (EMSA, 2010). EMSA’s (2010) goals are to reduce

the risk of maritime accidents, decrease marine pollution from ships and to prevent the loss of

human life at sea. In order to achieve this, important key tasks are: strengthening the Port State

Control regime; conducting audits of the Community-recognized classification societies, e.g. Lloyd’s

Register; developing common methodology for the investigation of maritime accidents; and

establishing a community of vessel traffic monitoring and information systems (EMSA, 2010).

2.1.3 The maritime domain in Sweden

There are several organizations tied to and relevant for reporting systems in the maritime domain in

Sweden. The three organizations that are deemed most relevant will be mentioned in this study: The

4

Swedish Maritime Administration (SMA), The Swedish Transport Agency (STA) and the Swedish

Shipowners’ Association (SSA).

The Swedish Maritime Administration (SMA) is a public enterprise within the transport sector.

SMA's primary tasks (Sjöfartsverket, 2010) concern the responsibility to provide such infrastructural

services as safe and accessible fairways that ensure the needs of shipping. These tasks include

supervising and maintaining breadth and depth of fairways as well as supervising, operating and

performing maintenance work on fairway facilities such as beacons, buoys and spar buoys

(Sjöfartsverket, 2010).

The Maritime Department of the Swedish Transport Agency (STA) was created in 2008 and inherited

many functions from the former Maritime Safety Inspectorate (Transportstyrelsen, 2010). The

department formulates regulations as well as examines and grants permits. It also has a supervising

role for the traffic in Swedish waters and works furthermore to improve safety as well as to prevent

negative environmental influence at sea. One important part of the department’s safety work is to

analyze maritime accidents and near-misses of merchant and fishing vessels (Transportstyrelsen,

2008).

The Swedish Shipowners’ Association (SSA) is a trade organization for the Swedish shipping

companies. SSA aims to support the Swedish maritime profession and trade, as well as to make

Swedish companies competitive and attractive on the international and national market both. SSA is

actively working with its members to create fair and equal competition and to promote and enhance

the work in such areas as maritime environment and maritime safety (Sveriges Redareförening,

2010). SSA’s work creates the possibility of sharing competence in various sectors of the maritime

trade as is achieved with committee work and by meetings conducted among the organization’s

members. SSA is also one of the initiators of the Swedish reporting system INSJÖ that will be

introduced in more detail in 2.3.2.

2.2 Definitions of accident, incident and near-miss There are several definitions of accidents, incidents and near-misses. The definitions introduced

below function as an underlying framework for how these three concepts are defined in this study,

see 3.3.

Bird and Loftus (1976) define incidents as undesired events that could (or do) result in loss (or

downgrade the efficiency of the business operation). Accidents are defined as undesired events that

results in physical harm to a person or damage to property. The authors also assert that accidents

usually are the result of contact with a source of energy above the threshold limit of the body or

structure.

Bird and Loftus (1976) present an accident-ratio study that analyzed 1 753 498 accidents reported by

297 cooperating companies. The triangle, as shown in figure 2, represents accidents and incidents

reported, not the total number of accidents or incidents that actually occurred. According to this

point of view near-miss is not a category of its own but rather a type of incident that is frequently

referred to as near-miss accidents (Bird & Loftus, 1976).

5

Figure 2. The figure shows the 1-10-30-600 ratio presented by Bird and Loftus (1976). For every 600 near-miss accidents,

there will be 30 property damage accidents, ten minor injuries and one serious or disabling injury

Cambraia, Saurin & Formoso (2010) adopt the concept of near-miss as an event separated from the

incident notion. The incident term is rather used as an umbrella term referring to any situation

where there is lack of safety - including accidents, near-misses and unsafe acts and conditions

(definitions of unsafe acts and conditions are not within the scope of this study and will therefore not

be further elaborated on). They define near-misses as instantaneous events involving the sudden

release of energy that has the potential to generate an accident, even though the consequences in

that case do not result in personal injuries or material damage, but usually in the loss of time. Near-

misses thus differ from accidents in the sense of not having any negative outcomes, such as damage

or injury. A near-miss could be the unfortunate event of a vehicle with engine failure stuck in a

railway crossing. A situation like this could lead to a dire outcome, but in this example due to some

fortunate turn of events, the train driver had enough time to brake and halt the train before impact.

This helped to prevent an otherwise inevitable collision. The event as it played out would most likely

have resulted in loss of time and a train schedule running late but hopefully not in personal injury or

material damage, thus making it a near-miss rather than an accident. This description of near-misses

is similar to the notion of incidents and the downgrade of the efficiency of the business operation as

described by Bird & Loftus (1976).

Heinrich, Petersen & Roos (1980) define accidents in another manner. They assert that it is

unnecessary and misleading to talk about accidents in terms of severity (e.g. a minor or major

accident) and do not use the term incident at all in their book Industrial Accident Prevention. Instead,

the potential effect or injury that stems from the accident should be graded in terms of severity; as

major injuries, minor injuries and no-injury accidents. Heinrich et al. (1980) define these three types

of injury as:

1) A major injury is any case that is reported to insurance carriers or to the state compensation

commissioner.

2) A minor injury is a scratch, bruise or laceration such as is commonly termed a first-aid case.

6

3) A no-injury accident is an unplanned event involving the movement of a person or an object,

ray or substance, having the probability of causing personal injury or property damage.

(1980, Heinrich et al., p. 63).

Even though the concept of injury is in the focus, a division into the three injury types above does not

seem to be fundamentally different from a division into the accident, incident and near-miss

category in terms of severity. No-injury accidents still share the fundamental features with the

incident notion as defined by Bird and Loftus (1976) and the near-miss concept as defined by

Cambraia et al. (2010). Heinrich et al. (1980) estimate that in a unit group of 330 accidents of the

same kind and involving the same person 300 accidents result in no-injury accidents, 29 accidents

result in minor injuries and one accident result in a major injury.

Figure 3: The 1-29-300 ratio, the foundation of a major injury (Heinrich, Petersen & Roos, 1980)

This pyramid like ratio has often been interpreted as- much like the triangle ratio of Bird and Loftus

(1976) depicted in figure 2 - to imply that the causes of frequency are the same as the causes of

severe injuries. In the fifth edition (the first edition was published 1931) of Industrial Accident

Prevention the authors (Heinrich et al., 1980) refute those previous notions and comment on what

has been written in earlier editions of the book when they state:

“Our original data of 1-29-300 were based on ‘accidents of the same kind and involving

the same person.’ The Figures are averages of masses of people and all kinds of

different accident causes and types. It does not mean that these ratios apply to all

situations. It does not mean, for instance, that there would be the same ratio for an

office worker and for a steel erector. (..)It also does not mean, as we have too often

interpreted it to mean, that the causes of frequency are the same as the causes of

severe injuries.“ (1980, Heinrich et al., p. 64).

Heinrich et al. (1980) acknowledge that the ratio might be problematic to use as a general model of

how accidents, incidents and near-misses interrelate. Different situations, environments and types of

accidents do not necessarily adhere to this model, as have been implied by statistic figures presented

after the claims of the 1-29-300 ratio (Heinrich et al. (1980). The view that frequently occurring no-

injury accidents or minor injuries lead to major injuries in a straightforward manner is thus being

challenged. Heinrich et al. (1980) further describe the problems stemming from viewing the 1-29-300

ratio as applicable in a broad sense. The interrelation between each step of the triangle seems not to

7

be as direct as the figure itself implies and the circumstances that produce different types of injuries

differ amongst each other; this leaves the authors with a new conclusion:

(..)”We have typically believed a 1-29-300 ratio, believed it might apply to all kinds of

accident types and causes, and then seen national figures. (..) that show that different

things cause severe injuries than the things that cause minor injuries. Obviously then

there are different ratios for different accident types, for different jobs, for different

people, etc. (..) This very difference might lead us to a new conclusion. Perhaps

circumstances which produce the severe accident are different from those that produce

the minor accident.“ (1980, Heinrich et al., p. 64).

The authors themselves realize that separate ratios might apply to different jobs and that the

circumstances that produce severe injuries possibly are not the same as those that produce minor

injuries. This open up for further discussion regarding the usefulness of these types of ratio models

and will therefore be discussed further in 6.2 of the discussion chapter.

Perrow (1999) introduces another perspective on accidents and incidents. He defines accidents and

incidents from a systemic view, where the complexity and interactive features of a socio technical

system as a whole are emphasized as important to be able to understand why accidents and

incidents occur. He divides a system into four different levels: (1) the whole system - e.g. a nuclear

plant, (2) a subsystem, e.g. a nuclear plant’s secondary cooling system, (3) a unit of parts, e.g. a

collection of parts that make up a steam generator or (4) a single part, e.g. a lone valve. He defines

these four levels and accidents versus incident as:

“An accident is a failure in a subsystem, or the system as a whole, that damages more than

one unit and in doing so disrupts the ongoing or future output of the system. An incident

involves damage that is limited to parts or a unit, whether the failure disrupts the system or

not.” (1999, Perrow, p. 66).

The major difference between accidents and incidents is thus the scope of the damage done (Perrow,

1990). An accident should according to this definition damage more than one unit of parts within the

system. As a result, the whole system or a large subsystem within it will fail. An incident is limited to

damage to single parts or to a unit of parts, even though the potential disruption could befall the

system as a whole.

Dekker (2010) presents a stance on accidents from a systemic point of view. An accident is an

emergent property in a system and a result of the system’s components doing their ‘normal’ work.

He describes system accidents as something that is possible even though each part of the system

seems to function normally, e.g. everybody are following local rules, common solutions and habits.

Dekker (2010) among many other proponents of a systemic view refutes the idea of simple, linear

and causal explanations to complex situations (or negative events, such as accidents and incidents)

where a multitude of relations and interconnections coexist within the system. Dekker (2010)

underlines that this linear and casual view of accidents stems from an objectivistic point of view,

where events and accidents can be broken down to simpler and more understandable parts. He

furthermore warns that this creates the need to search for and to blame the ‘single’ part or parts

that stand as the root cause of the accident.

8

2.3 Reporting systems The use of reporting schemes is becoming widespread in domains such as the chemical process

industry, transportation and health care (Schaaf & Kanse, 2003). There are several studies on

incident reporting and near-misses in these domains (Jones et al., 1999; Lawton & Parker, 2002;

Evans et al., 2004; Elder et al. 2006; Sanne, 2008; Cambraia et al., 2010; Barach & Small, 2000).

Johnson (2003) divides reporting systems into three main categories; open, confidential and

anonymous. Open systems provide all details concerning the report, in confidential systems are

identification only available to alleged responsible parties, whereas anonymous systems de-identify

and often to some degree de-contextualize stored reports. Johnson (2003) furthermore ascribes

levels, in terms of local, national or international usage, to reporting systems. These characteristics

imply different strengths and weaknesses (Johnson, 2003). An open system risks being limited in its

use if the users are afraid of punishment and unwelcome exposure in the media. A benefit with an

open system is that an investigator has all collected information available. A confidential system

builds on trust, in the sense of that the ‘responsible parties’ that have access to all information do

their job properly. Confidential computer-based online reporting systems might create new security

issues and feel less trustful to people not used to computer-based systems. Anonymous systems

might give the reporter more confidence in their submission, though an apparent problem might be

the risk of decline in quality when the accountability of the submitted reports is removed. Johnson

(2003) mentions the paradox of anonymity in reporting systems. He presents an example from the

aviation domain where many people emphasize the importance of anonymity at the same time as

they acknowledge that full anonymity requires de-contextualized reports. The vital information that

could benefit an accident investigation might at the same time be part of the context and pose as

identification. The removal of this information could render a report much less useful. Johnson

(2003) also mentions that local reporting systems might tackle this problem better due to a smaller

scope and more inherent local context to that can be used in an investigation.

Critique against anonymity in reporting systems is also presented by Barach and Small (2000), due to

the potential threat to accountability and transparency. Barach and Small (2000) note that full

anonymity risks being counterproductive in the sense of that you cannot contact reporters to get

more and in some cases perhaps critical information. They also note that there is a risk that the

reliability might be lower when accountability is withdrawn.

The following sections will present reporting systems in the maritime domain, studies showing

barriers to reporting in general and studies acknowledging barriers that are found specifically in near-

miss reporting. A cross-domain overview of near-miss reporting is of relevance to learn more about

and perhaps find similarities and differences to near-miss reporting and its potential barriers within

the maritime domain.

2.3.1 Reporting systems in the maritime domain

Reporting accidents and near-misses at sea is mandatory and bound by legislation for Swedish

merchant and fishing vessels. This compulsoriness aims at supporting the authority when deciding

whether legal action should be taken as well as to help the responsible authority to prevent further

accidents (Transportstyrelsen, 2009). It is each ship’s master or ship owner that is responsible to

report these events (Transportstyrelsen, 2009). Accidents and near-misses are reported on the form

9

“Report on Accidents at Sea” that is sent to the Maritime Department of the STA by mail

(Transportstyrelsen, 2009). These types of reports are common in most professional industries

(Zachau, 2008) and are often stored in computerized databases, such as the Swedish maritime

database SjöOlycksSystemet (SOS) - which will be presented later.

Reporting systems can be mandatory by law as well as non-mandatory to partake in (Barach & Small,

2000). Systems that are mandatory often have a larger ratio of accidents whereas non-mandatory

systems often offer confidentiality and strive to stimulate near-miss reporting, generating reports of

events that otherwise might get unnoticed during accident prevention work (Zachau, 2008).

One of the arguments for near-miss reporting is the ‘iceberg’ shaped ratio (Jones, Kirchsteiger and

Bjerke, 1999; Heinrich et al., 1980; Bird & Loftus, 1976) - see 2.2 for figures and further definitions -

which implies that near-misses at the base stand in direct connection to the amount of incidents and

accidents further up the iceberg. Other benefits of near-miss reporting include a more proactive

approach to safety work (Barach & Small, 2000).

The IMO's guidance on near-miss reporting (2008) states that every company should investigate near-misses as a regulatory requirement, as mentioned in the ISM code - and further define near-misses as a sequence of events and/or conditions that could have, but did not result in loss (such as human injury, environmental damage or negative business contact). The IMO (2008) further states that to gain full benefit of near-miss reporting both seafarers and onshore employees need to understand the definition of near-misses. The IMO (2008) also mentions explicitly that companies must be clear about how reporters and the persons involved will be treated when a report is made and in which circumstances the reporter and those involved will be guaranteed a non-punitive outcome and confidentiality. Each company should strive to create a just culture that is built on both trust and responsibilities, and where sharing or reporting essential safety-related information is made without fear of retribution.

One example of a confidential reporting program is for aviation and the maritime in the United

Kingdom (UK) the Confidential Hazardous / Human Factors Incident Reporting Programme (CHIRP).

The reporting system's maritime program has been operative since 2003 with the aim to contribute

to the enhancement of maritime safety in the UK, by providing an independent and confidential,

though not anonymous, reporting system for employees and associates within the maritime

industries (CHIRP, 2007). Reporting to CHIRP can be done both online through the website or by

sending an e-mail and through ordinary mail or by telephone / fax (CHIRP, 2007).

There are other maritime reporting systems, such as The Mariners' Alerting and Reporting Scheme,

MARS. MARS is a confidential reporting system, with the possibility to be anonymous, run by The

Nautical Institute in London. The Nautical Institute functions as an international organization and

forum for qualified seafarers and others with an interest in nautical matters (MARS, 2008). The

objectives of the reporting system are to allow reporters to report accidents and near-misses without

being afraid of litigation and to exchange information so that valuable lessons may be learnt by

others; which might help to prevent similar accidents in the future (MARS, 2008). The reporter, often

a member of the Nautical Institute is guaranteed anonymity for himself as well as for the ship.

Reports are sent online through the website or printed and sent through ordinary mail. The reports

are published on the Nautical Institute’s website as well as in their monthly journal (Zachau, 2008).

10

In Sweden, SjöOlycksSystemet (SOS) is a database for accidents and near-misses aimed at the

Swedish merchant fleet. SOS was at the time of Zachau’s (2008) study operated by the Swedish

Maritime Safety Inspectorate (Zachau, 2008). The reports are sent in by ship captains or companies

as legislation demands (Zachau, 2008), though Zachau (2008) noted that only 7-8 percent of the total

reports are categorized as near-misses. The information in the database is public. Even though it

does not contain the names of any persons, other information like ship names, positions and date

make identification possible (Zachau, 2008).

According to Zachau (2008), the above mentioned low numbers of reported near-misses led to an

agreement among the Swedish Maritime Safety Inspectorate (Part of the Swedish Transport Agency

since 2009) together with ship owners, employees and the maritime industry to create the

autonomous, confidential and anonymous database INSJÖ, with the goal to remedy the inaccurate

ratio between accidents and near-misses. Section 2.3.1 has introduced reporting systems in the

maritime domain, both systems primarily at work in Sweden and systems that are used

internationally. The presented reporting systems range from open to anonymous and are aimed for

both accidents and near-misses. Section 2.3.2 will present the reporting system INSJÖ in more detail.

2.3.2 INSJÖ INSJÖ is an autonomous, confidential and voluntary reporting system with web-based reporting

forms. Reports stored in the database can be retrieved directly online. The database is not open to

the public and you need to login to view the reports (Zachau, 2008). INSJÖ’s aim is to follow the ISM

code, as adopted by the IMO, see 2.1.1.

Involved parties in the INSJÖ development are the Maritime Department of the Swedish Transport

Agency (STA), the Swedish Shipowners’ Association (SSA) and the Swedish Maritime Agency (SMA).

The Merchant Marine Officers' Association, Swedish Ship Officers' Association, Swedish Seamen's

Union and other concerned unions have also been involved in the project (INSJÖ, 2007).

INSJÖ contains roughly 2500 reports (INSJÖ, May 2010), with approximately 300 new reports added

annually (Zachau, 2008). The essential content of the database are reports from companies, ships,

safety committees and crews on board Swedish ships (INSJÖ, 2005).

The term near-accident is used in favor of the term near-miss in the INSJÖ database (INSJÖ, 2007),

even though the properties of the first term coincide with the near-miss definitions in 2.2. For sake of

coherence, the term near-miss will still be used in section.

The DPs for each shipping company have the possibility to forward the reports from their company

to INSJÖ, and thus share knowledge of accidents, near-misses and non-conformities (non-

conformities will not be presented further in this study) nationwide to all participants in the INSJÖ

collaboration. It is not obligatory for shipping companies to do near-miss reporting through INSJÖ

even though near-miss reporting in some form is strongly encouraged and closely tied to the DP role

as it is assigned by legislature.

A report sent to INSJÖ is written by a reporter (in most cases the DP) on the basis of categorizations

used in INSJÖ’s reporting form, e.g. type of ship, type of event (accident, near-miss or non-

conformity), event description, the cause of the event, the consequences of the event and measures

11

taken afterwards. All information regarding the event itself is written down in a free-text format. The

independent company in charge of the database categorizes these answers in order to make the

report searchable in the database. The DP gets feedback on his or her reports (originally sent in from

a reporter on one of the shipping company’s vessels), in the form of similar cases stored in the

database. This feedback can be used to guide the DP when proper corrective actions are decided

(INSJÖ, 2005).

This feedback process is shown in figure 1 below.

Figure 1. The feedback process in INSJÖ. (Copyright ICC, IPSO Classification & Control AB ICC, IPSO Classification &

Control AB, Retrieved November 2009.)

Zachau (2008) did an analysis of INSJÖ and compared the voluntary INSJÖ database with the public,

SOS database (SjöOlycksSystemet) that does not provide anonymity and that cannot guarantee that

legal actions will be excluded. Voluntary and confidential databases like INSJÖ should, according to

several studies mentioned by Zachau (2008), contain a higher ratio of reported near-misses

compared to incidents. He found that INSJÖ did not contain the expected ratio of near-misses in the

database. The ideal relation would be 1:100, which would give more power to conduct such tasks as

proactive safety work, due to a large amount of analyzable near-miss events, whereas INSJÖ had only

a 50:50 relation. This is still a step in the right direction, according to Zachau (2008), if compared to

SOS that contains far less near-miss reports in relation to the number of accidents. This most likely

stems from the fact that accidents, by definition (see 2.2), often are easier to recognize, harder to

ignore, due to their negative outcome, and obligatory to report. It is in contrast harder to always

correctly identify and make sure that near-misses are reported in the same manner.

2.3.3 Near-miss reporting

Barach & Small (2000) mention several advantages using near-misses in reporting systems. They note

that near-misses occur 3-300 times more often than negative events, such as incidents, which makes

a quantitative and statistical analysis possible, this might help identify patterns in the data (Johnson,

2003). Barach and Small (2000) also note that the study of strategies and mechanism for making

recoveries - that might determine whether the outcome will be negative or not - enhances proactive

12

means to hinder accidents. They furthermore mention that the post accident / incident hindsight bias

– the inclination to rate a phenomenon as more predictable than it actually is - can be reduced when

studying the interrelation between accidents and near-misses.

Jones et al. (1999) mention the ‘iceberg‘ relation between the numbers of near-misses, minor

incidents and major accidents as has been demonstrated in earlier studies (Heinrich et al., 1980; Bird

& Loftus, 1976) and depicted by Heinrich et al. (1980) and Bird and Loftus (1976), see 2.2. Reducing

near-misses at the ‘bottom’ of the iceberg will supposedly affect and reduce the amount of incidents

and accidents further up.

Jones et al. (1999) point out that the actual amount of reported near-misses is far from satisfactory in

many domains, and most likely not even near the actual amount or level of near-misses that occur in

reality. This suggests that an increase of near-misses in different kind of incident reporting systems

can and should be seen as a positive indicator of safety performance in the sense of that the near-

miss reporting gets stimulated and helps to unveil occurrences of near-miss events that are not

reported at present. Jones et al. (1999) present Norsk Hydro and their focus on near-miss reporting

as an example where it was evident that the number of accidents lowered when the near-miss

reporting went up. They suggest that the rate of near-miss reports is an important numerical

indicator of industries’ safety awareness. The term safety awareness is not further explained or

defined by the authors and will therefore not be elaborated upon in this study.

2.3.4 Barriers to Reporting

A collaborative hospital study (Evans, Berry, Smith, Esterman, Selim, O’Shaughnessy, & DeWit, 2004)

showed that self-perceived barriers to incident reporting - near-misses included - for both doctors

and nurses were lack of feedback and organizational factors relating to structures and processes for

reporting (e.g. inadequate feedback on actions taken, long forms and insufficient time to report).

Almost two thirds of all respondents in the study believed that the above-mentioned lack of

feedback was the greatest deterrent to reporting.

Van der Schaaf & Kanse (2004) highlighted differences in perceived reasons for not reporting

incidents in the chemical process industry. The management and safety staff did to some extent

anticipate fear and shame as potential barriers to operators. They also anticipated that operators

would view often experienced and common risks as something negligible to report, in the sense of

that common occurring events would be viewed as ‘nothing new’, widely known by the personnel

and without learning potential. Successful recoveries were also anticipated to be viewed as

superfluous to report by the operators, because the situation would likely be seen as taken care of.

To the surprise of management and safety staff the study showed a genuine difference between

some of the anticipated barriers mentioned beforehand and the one brought up by participant

operator. The operators de-emphasized fear and shame as barriers contrary to the beliefs of the

management. The barriers mentioned the most concerned the fact that no remaining consequences

were to be found, which made reporting non-valuable and insignificant. Other barriers were labeled

as not applicable and referred to various reasons such as miscommunication and administration

errors.

Elder, Graham, Brandth and Hickner (2007) studied barriers and motivators for what they present as

error reporting (reporting of events that could lead to incidents or accidents) within the domain of

13

family medicine in the US. The term error reporting will not be further used or elaborated upon in

this study. Common themes found during several focus groups were: 1) Burden of effort, 2) Clarity of

request, 3) Perceived benefit and 4) Properties of the error.

The burden of effort in reporting referred mostly to lack of time to report and a risk of forgetting to

file a report at all. The clarity of request referred to the difficulties to know what to report.

Repetitive and frequent errors were found in this category as well as errors that were unlikely to

recur. Other barriers in this category concerned what to write in the report and if it applied to a

certain person’s job to report a particular event. The benefit of reporting was not acknowledged or

seen as a job requirement by some of the participants. Certain properties of an error also related to

barriers; errors that were deemed as not serious and errors that were self-made were less reported.

Motivators to reporting were in most cases found in inverse of the barriers though more scarcely

mentioned by the participant groups. Other common motivators mentioned involved receiving some

sort of perceived benefit such as feedback or knowledge that lessons were made known to

colleagues. Anonymity was also seen as a motivating factor in making reports.

In his study of the Swedish railroad domain, Sanne (2008) highlights how different accident

etiologies and discrepancies between official policies and local practices can conflict in ways that

hinder the official incident reporting process. Sanne (2008) describes that the reason for non-

reporting in the railroad domain is due to different accident etiologies. Occupational and informal

storytelling is the occupational norm, while the official incident-reporting scheme is not. Telling

stories to teach and learn from each other can function as a way to address risk, though from a more

narrow and local perspective; stories are shaped by the shared values and norms within the social

context in which they are told, and the value of storytelling risk to be too limited in a larger

organizational perspective. Sanne (2008) mentions how an awareness of these limitations could

create insights in how change to a better incident-reporting climate could be accomplished and what

kind of obstacles that has to be overcome. One of the most important conclusions from Sanne’s

(2008) fieldwork is that, in order to make incident-reporting work properly, employees must be given

ownership of the incident reporting system, and know how and why to use it. Fear of disciplinary

action must be addressed and a better focus on finding root causes, as often lacking in occupational

storytelling, is important; as well as giving more feedback and education in the principles of incident-

reporting systems within a more systemic perspective.

2.3.5 Barriers to near-miss reporting

Barriers relating to near-miss reporting are described in the IMO’s guidance of near-miss reporting

(2008). Common barriers mentioned are fear of being blamed, disciplined, embarrassed or found

legally liable. Organizational barriers are also mentioned, such as unsupportive company

management attitudes, insincerity about addressing safety issues and discouragement of the

reporting of near-misses by demanding that seafarers conduct time consuming investigations in their

own time. The IMO (2008) states that these barriers can be overcome by initiatives from the

management. This can be achieved by encouragement of a just culture approach which covers near-

miss reporting (IMO, 2008). The culture should be just in the sense of that the company gives people

responsibility, earn their trust and promote that sharing sensitive information in most cases do not

bring negative consequence to the people involved. The IMO (2008) describes the just culture as

14

featuring an atmosphere of responsible behavior and trust where people get encouraged to report

important safety-related information without fear of reprisals. Even though a just culture is present

in a company, the IMO (2008) emphasizes that a distinction between acceptable and unacceptable

behavior must be upheld. They furthermore state that unacceptable behavior will not go unnoticed

or be without the risk of facing consequences.

The just culture concept also includes supplying confidentiality to reporters when reporting near-

misses, to ensure that enough resources are given to the investigation at hand and that near-miss

reporting gets followed through with suggestions and recommendations for future conduct (IMO,

2008).

Barach and Small (2000) draw conclusions from domains (though not the maritime one) where

reporting near-misses have been institutionalized to gain more insights to help to create similar

schemes in health care, insights that might enhance the reporting and the prevention of medical

mishaps. They list domains such as aviation, nuclear power technology, petrochemical processing,

steel production and military operations to have these kinds of near miss reporting schemes. The

authors list several barriers that were found in 12 non-medical incident reporting systems.

The authors divide different kinds of barriers (and incentives to those barriers) - found in the various

studied domains - into three main categories: individual, organizational and societal; where each

larger category could be further divided into four subcategories or aspects: Legal, cultural, regulatory

and financial, see table 1 below.

Table 1. Barriers and incentives to reporting

Individual Organizational Society

Legal

Barrier Fear of reprisals, lack of

trust.

Fear of litigation, costs,

sanctions, undermine trust,

bad publicity

Legal impediments to peer

review, confidentiality, and

multi-institutional databases

Incentive Provide confidentiality

and immunity

Provide confidentiality and

immunity

Ensure accountability, enforce

reporting statuses

15

Cultural

(values,

attitudes,

beliefs)

Barrier Dependent on

profession, code of

silence, fear of

colleagues in trouble,

skepticism, extra work

Dependent or organization,

pathological, bureaucratic,

generative cultures, don’t

want to know

Wide public trend towards

disclosure, lack of trust owing to

highly publicized medical errors,

concerns that professions are

too privileged, lack of education

about system effects

Incentive Professional values:

philanthropic, integrity,

educational, cathartic

Become a leader in safety

and quality; good for

business

Enhanced community relations,

build trust, improve health care,

transparency

Regulatory

Barrier Exposure to malpractice,

premiums will go up,

investigation and

potential censure,

license suspension and

subsequent loss of

income

It doesn’t apply to us, we do

our own internal analysis

process, they can’t

understand our problems

anyway

Need more effective

regulations, resource intense

Incentive Prophylactic, follow the

rules

Fear of censure Enhances regulatory trust, more

public accountability

Financial

Barrier Loss of reputation, loss

of job, extra work

Wasted resources, potential

loss of revenue, patient care

contracts, not cost effective

Cost more tax dollars to

enforce, more bureaucracy

Incentive Safety saves money Publicity relations, improve

reputation of quality and

safety

Improves confidence in

healthcare systems

Table 1 present barriers and incentives to reporting in 12 domains. Legal, cultural, regulatory and financial subcategories

are viewed through their impact on the individual, the organization or the society

Barach and Small (2000) found that disincentives to reporting was the extra work needed,

skepticism, lack of trust, fear of reprisals and lack of effectiveness of present reporting system.

Incentives to report would be confidentiality, some degree of immunity, and that the reporting

system should be philanthropic (that reporters identify with patients and other healthcare providers

that benefit from the data), and therapeutic (in the sense of that reporters learn from reporting

about adverse advents). Barach and Small (2000) mention several important factors that determine

the quality and success of incident reporting systems. These include having an independent

outsourcing of the report collection, analyzing reports with help of peer experts, having sustained

leadership support, making it easy to report and supplying rapid meaningful feedback to reporters

and all interested parties.

16

Barach and Small (2000) highlight several changes or conflicts that can occur when taking the near-

miss perspective. If the focus changes from errors and adverse events, the near-miss perspective

might move the focus to resilience, in the sense of that successful recoveries from accidents are

emphasized. There might also be tradeoffs between large aggregate databases and more regional

systems. A national system might help to capture more rare events where more regional ones

instead provide more specific and local feedback more efficiently.

17

3 Aim This study’s aim is to examine near-miss reporting and to identify barriers that might hinder or limit

near-miss reporting in the Swedish maritime domain. The study focus on both near-miss reporting

through the national INSJÖ database - that is used by several Swedish shipping companies to gather

and share knowledge about incidents and near-misses - and through the obligatory and local near-

miss reporting between ship and office. Near-misses are in this study defined as events that could

but did not result in damage or injury but otherwise share aspects with an accident (Cambraia et al.

2010).

3.1 Research question

What kind of barriers against reporting in general and near-miss reporting in particular can

be found prevalent in the Swedish maritime domain?

3.2 Boundaries of the study This study aims to examine and identify barriers prevalent in the Swedish maritime domain, and only

shipping companies with ships flagged in Swedish waters participate in the study. Not all types of

Swedish shipping companies are represented in this study.

Barriers highlighted in this study are the ones found when analyzing the interviews with selected

representatives, seven DPs and four officers, from shipping companies in Sweden. The study does

not focus on other parties’ view of barriers, such as other company personnel or representatives

from other organizations. Specific details concerning individuals, ships, companies or events are not

brought up in this study.

3.3 Scope of this thesis Accidents in this study are defined as all safety related events that are accompanied by negative

consequence, including incidents (Perrow, 1999), errors and other adverse events as mentioned by

Barach & Small (2000); this is in contrast to near-misses that do not result in injury or property

damage (Bird & Loftus, 1976).

18

4 Method Ethnographic methods are about studying and getting insights of the daily lives of others (Emerson,

Fretz & Shaw, 1995). The core activities in ethnographic field research are often participant

observation, as the ethnographer participates in the daily routines of the studied setting, and an

accumulation of written material of observations and experiences made (Emerson et al. 1995). This

kind of traditional field research is often time consuming and extensive. When looking at socio

technical systems, these potential drawbacks have led to tradeoff-methodologies or focused

ethnographies (Knoblauch, 2005) that balance time limitation and knowledge gain. A focused

ethnography, in contrast to conventional ethnographies, often means conducting shorter field visits

and having a more extensive data collection during these visits, as might be appropriate when time is

an important factor or when the setting under scrutiny is one under change (Knoblauch, 2005). To

gain more knowledge on one or more specific aspects of daily work in the maritime domain, such as

the view on reporting and potential barriers that might hinder reporting in near-miss events,

qualitative methods can be used to capture ’life as it is lived’ (Boeree, 2010). This can be done by

studying results of past living, observing the present or eliciting data by the use of methods such as

interviewing or conducting focus groups. These methods let the researcher ask questions and, in the

case of focus groups, initiate tasks that encourage a group to discuss and labor on specific topics

(Boeree, 2010).

The material analyzed in this ethnographically inspired study has been collected by conducting semi-

structured interviews - which was deemed the best choice time and resource wise - with deck

officers and DP personnel. This chapter will begin with a presentation of interviewing as a qualitative

method in 4.1 and continue with a description of the data collection and data analysis in 4.2 and 4.3.

Lastly, 4.4 describe how validity, reliability and objectivity concepts can be applied when using

qualitative methods.

4.1 Interviews as method Interviews are a common and versatile way to learn more about certain topics of interest and to

capture and investigate participants' attitudes (Jordan, 1998). Interviewing ranges from highly

structured interviews with all questions developed beforehand to completely unstructured ones

(Benyon, Turner & Turner, 2005). Unstructured interviews contain open-ended questions that give

the respondents the opportunity to steer the interview in preferred directions. This might be

beneficial when explorative studies are conducted, when the background information is limited (as

when the interviewer does not know what the issues of concern are) or when the interviewer lacks

domain knowledge (Jordan, 1998). Structured interviews contain a pre-set collection of questions,

which might be a good format if conducting a structured quantitative analysis where certain

variables are studied (Jordan, 1998). This type of interview also requires a clear idea of the issues of

interest (Jordan, 1998).

The semi-structured interview uses questions prepared beforehand without losing the opportunity to

explore new topics as they are brought forth during the interview. The interviewer should have

considered relevant topics and have expectations of what type of questions that might elicit relevant

responses from the respondents beforehand (Jordan, 1998).

19

Jordan (1998) mentions that interviews have certain advantages to other means of collecting data

such as questionnaires, as they share certain aspects with interviewing but often are more static.

When interviewing, the contact with the respondent is direct and misinterpretations can be repaired,

for example by rephrasing or providing more context. This in a way that ordinarily is not possible

when reading text only and the respondent has to make an interpretation based solely on the

information written down. Jordan (1998) also emphasizes that it is easier to compensate for

potential deficiencies like ambiguity that risk creating confusion and misconception in written text,

this due to the two-way communication during the interview session.

Interviewing as a qualitative method is not, as any method, without potential drawbacks. Problems

might occur if the interview is conducted without preparation or if the interviewer biased. How a

question is asked might affect how a respondent will answer that certain question (or how the

interviewer chooses to interpret the answer given) (Boeree, 2010). Section 4.4 presents how

concepts for judging quality of one’s findings, e.g. validity, reliability and objectivity can be applied in

qualitative research.

4.2 Data collection To collect material, on the near-miss reporting and its potential barriers in the Swedish domain,

semi-structured interviews with seven shipping companies were held. A literary overview of near-

miss reporting in other domains were conducted beforehand, this overview helped me formulate

topics and questions (to be included in the interview template) deemed relevant for identifying

barriers that might be prevalent in the maritime domain as well as in other domains.

To get perspectives from both the office and from onboard the ships I interviewed both DP personnel

and deck officers at sea. The participating companies were also divided between companies that

report to INSJÖ and companies that do not, this to make potential differences in perspective salient.

Before the interviews could be conducted, the DPs of several Swedish shipping companies were

contacted by phone. The companies called where chosen based on convenience and location. They

were given information regarding the study and were asked if they could participate, and if it was

possible to interview someone who regularly made reports from one of the company’s ships; in

those cases this was possible the DP supplied contact information to selected officers, in charge of

the onboard reporting. All of the interviewees were men.

The interviews were semi-structured (Benyon et al., 2005). An interview template (See Appendix A)

was used but not followed strictly Follow-up questions were given when unforeseen topics arose.

This generated new questions and themes not thought of beforehand. The template was created to

help cover all themes that were deemed interesting. There were four versions of the template with

slightly different questions; the groups considered were INSJÖ-DP, INSJÖ-Officer, Non-INSJÖ-DP and

Non-INSJÖ-Officer. The following broad themes were judged relevant to investigate further.

1) The view on near-misses

2) The view of reporting

3) The view of DP

4) The relation between crew and DP

5) Potential barriers to reporting

20

6) Feedback when reporting

7) The view on responsibility onboard

Theme 7 was specific to the template given to the officers, to learn more about the view on

responsibility onboard from the reporting officer.

Not all the template questions (See Appendix A) were asked if I deemed them redundant or already

answered by the interviewee. One interview could not be given in person or by phone due to

constraints regarding time and resources. This participant answered similar questions by e-mail

instead; this interview therefore lacked the semi-structure of the other interviews and was e-mailed

as a version of the interview template, edited to suit the e-mail format better.

Each participant was given a short introduction regarding the aim of the interview and the study as a

whole. Each interview in person or by phone was recorded after the respondent had given his

consent to participate. All participants were briefed about being anonymous, about that they could

abort the interview at any time and about that all material recorded is handled strictly confidential,

in line with ethical principles when interviewing (Bryman, 2002). The participants were also informed

that the gathered material would not be used for any commercial purposes. The final draft of the

thesis will also be e-mailed to each participant before the study goes into print, this to let the

respondents comment on the material and conclusions made.

The interviews were conducted with representatives of seven different shipping companies, four of

them were tanker companies and the other handled other types of cargo. Five of seven companies

report to INSJÖ. Interviews were conducted with seven people in the offices ashore - in all cases but

one interviews were made with the person designated by regulation to have responsibility over

safety management (this group will still be described as DPs) - as well as four nautical officers

onboard ships belonging to four of the total seven selected companies. Five out of seven interviews

with DP personnel were conducted in person while the other two were interviewed by phone. The

one participant that did not work as DP was second in command of the company’s safety work and

had good insight in the reporting matter.

Of the four interviews that were held with officers, only one interview was made onboard a ship (due

to limited time and travel resources), two interviews were made by phone and one was answered by

an officer through e-mail due to limited communication possibilities at sea. Three out of four officers

belonged to shipping companies using INSJÖ to report near-misses.

4.3 Data Analysis After the interviews were completed, the material was transcribed and color coded according to the

different themes found during this same process. The transcriptions were verbatim, though not on a

very detailed level and only focused on what the informants were saying. The transcriptions did not

include overlaps, pauses or potential communicative non-verbal gestures such as laughs, nods or

humming in agreement. The material was used in two separate analyses; the first analysis (analysis

1) is described in 4.3.1. This analysis was done in a naturalistic fashion, with highlighted categories

deemed relevant chosen by me and without the use of any particular underlying method. The second

analysis (analysis 2) is described in 4.3.2 and was done after the completion of the first analysis. It

21

uses the categories from by Barach and Small’s (2000) presentation of barriers and incentives to

near-miss reporting in various domains, see table 1 in section 2.3.5.

4.3.1 Analysis 1

To make the material more manageable and lucid, the highlighted themes identified in the material

were plotted in a mind map like manner. This mind map includes two parts, one part that illustrates

the material from the interviewed DP personnel and another part that depicts the material from the

interviewed officers. Each participant’s highlighted statements from the material were transformed

into subcategories and organized according to each theme in the mind map, see figure 2 and 3. This

categorization made it possible to group participants’ quotations together and make differences and

similarities between the different participants visible; and to discern if there were differences among

the four groups: 1) INSJÖ-DP, 2) INSJÖ-Officer, 3) Non-INSJÖ-DP and 4) Non-INSJÖ-Officer. The two

visualizations can be seen in figure 2 and 3 below.

Fig. 2 Visualization of the eight themes, with subcategories, from the mind map generated from the four officer’s

statements. Each subcategory shows how many of the respondents that mention a particular topic

Figure 2 shows eight different themes that were used to categorize the subcategories that held

quotations from the officers. These themes were:

1. Why report near-misses?

2. The role of the DP

3. Barriers

4. Means to increase reporting

5. Feedback

22

6. Anonymity

7. External parties that affect reporting

8. INSJÖ

Fig.3 Visualization of the eight themes, with subcategories, from the mind map generated from the seven DP’s

statements. Each subcategory shows how many of the respondents that mention a particular topic

Figure 3 shows eight different themes that were used to categorize the subcategories that held

quotations from the DPs. These themes were:

1. Why report near-misses?

2. The role of the DP

3. Barriers

4. Means to increase reporting

5. Feedback

6. Anonymity

7. External parties that affect reporting

8. INSJÖ

Some changes in the presented themes were done during the course of analyzing the material, I

deemed one theme (the role of the DP) irrelevant to be a theme of its own, in regards to the aim of

the study. There were also other minor rearrangements and name changes to the themes. In some

cases parts of the gathered material were deemed relevant to more than one theme and will thus be

mentioned at more than one place in section 5.2 and 5.3 in the following analysis chapter. All

23

statements have been numbered to correspond to the representing DP or officer, this to make it

possible to see whether a representative has more than one statement in a subcategory.

The highlighted material was paraphrased (Purdue Online Writing Lab, 2010) from direct quotation

into my own words (Emerson et al. 1995) and then translated from Swedish to English. The

paraphrasing was made to make the translation simpler to handle, this due to the potential problems

of preserving the feel and accuracy of a direct translation from Swedish to English, for me as a non-

native speaker of the English language.

4.3.2 Analysis 2

Different categorizations highlight and mould the interpretation in various ways, and might lead to

new insights and conclusions from the same material. Table 2 in 5.5 divides the material into four

dimensions:

1. What the DP mentions as barriers and incentives on the individual level

2. What the DP mentions as barriers and incentives on the organizational level

3. What officers onboard mention as barriers and incentives on the individual level

4. What officers onboard mention as barriers and incentives on the organizational level

I analyzed the transcribed material once again from the beginning. Statements were once more

divided and categorized, though this time, the barriers and incentives listed by Barach and Small

(2000) were used as a template. The gathered material does not concern the societal category, which

was removed from the analysis. DPs and officers mention barriers on both the individual and

organizational level, which made it necessary to use the four dimensions presented above.

Respondents’ statements that I deemed to be of similar nature and meaning were grouped together

and generalized to create categories that would make the similarities and differences between the

two groups more salient, even though single statements of interest were highlighted as well.

4.4 Validity, reliability and objectivity When assessing the quality of one’s research, you often judge the work by using the concepts of

validity, reliability and objectivity. These kinds of tools for quality assessment are as important in

qualitative field research as in experimental design (Lützhöft, 2004). The use of above concepts (and

what they entail from experimental research and the natural sciences) might prove to be problematic

and poorly reflect methods often used in qualitative research (Lützhöft, Nyce & Petersen, 2009). This

section will present concepts used in the qualitative field that are of relevance when discussing

methodology in 6.3

When looking at the quality of one’s findings in qualitative research it is more appropriate to use the

term credibility rather than applying such validity terms as internal validity (the causal relation

between two variables) and construct validity (whether we actually measure what we think we

measure). This shift of concepts is a way to evade experimental concepts and the search for truth or

falsity of a proposition, a focus that ill befits qualitative research (Lützhöft et al. 2009). Credibility

concerns the degree to which a phenomenon is interpreted correctly given the data at hand and to

the soundness of the arguments given (Lützhöft et al. 2009). The goal of credibility is to show that

the link between the reconstructed world of the qualitative researcher and the respondents are

credible and phenomenological sound (Lützhöft, 2004). This could be accomplished through such

24

measures as a triangulation of methods, sources and investigators as well as with prolonged

engagement and through persistent observation (Lützhöft et al. 2009).

It is in the same manner more favorable to use the concept of transferability instead of external

validity in qualitative research. Transferability concerns whether it is possible to generalize, e.g.

observations and findings, to other domains, contexts and populations. Transferability is said to be

possible by providing such a rich or thick description that generalization can be derived from this

description (Geertz, 1973). Lützhöft et al. (2009) mention though that it is hard to assess when and

what kind of thickness that is sufficient in a given fieldwork situation.

The traditional definition of reliability (Heiman, 2001) is the extent to which a measured

phenomenon is the same every time it is measured in the same way. Testing of reliability by this

definition might make sense in a controlled experimental setting where randomness can be

minimized by design (Lützhöft et al, 2009). In real world settings this definition poorly reflect the

qualitative methodologies often used. Other definitions are therefore preferable (Lützhöft, 2009).

Two types of qualitative reliability are quixotic reliability and synchronic reliability (Lützhöft, 2004).

Quixotic reliability concerns the possibility that one single method misleadingly yields consistent

results. In qualitative research this might be misleading in the sense of that the information given

might be ‘rehearsed’. Synchronic reliability concern if observations are similar within the same time

period. If this is not the case these observations might give birth to insights of interesting potential

differences in different observer’s view of the world.

Objectivity, as traditionally viewed in the natural sciences, concerns knowledge’s independence from

all external influence and opinion (Lützhöft et al. 2009). Dekker (2010) discusses this world-view in

relation to its influence on human factors and the concept of human error (and to other fields in the

social sciences). He explains how this ‘Cartesian-Newtonian’ view implies that there are real

undeniable truths to discover about the world, bound to the laws of mechanics; and how complexity

is seen as readily reducible to smaller and more understandable parts if needed. This objectivistic

stance is problematic in qualitative methodology.

Objectivity in qualitative research is not so much about discovering the real undeniable truth, in the

above sense (Lützhöft, 2009). If one should apply this ‘Cartesian-Newtonian’ view when conducting

qualitative studies he or she would miss a crucial point of qualitative research entirely. Objectivity in

qualitative research concerns getting reliable insights in someone’s world-view, their perspectives

and the logic that goes with it (Lützhöft, 2004). This has nothing to do with finding the ‘truth’ in the

objective sense, in the same sense of that the this study does not concern finding some absolute

truth inherent in the respondents’ statements or a ‘true’ interpretation of this material. A certain

truth does not ‘cancel’ out everything else in qualitative research; diversity among findings is not a

negative thing but rather a source that gives you more insights of a phenomenon or let you gain a

deeper look into someone’s world-view (Lützhöft, 2004). Dekker (2010) furthermore highlights the

possible ethical complications that could arise when someone is an ‘adjudicator of truth’, the

researcher is right while the subjects (and possible their world-view) is wrong. This point of view

presented by Dekker (2010) goes against the qualitative research paradigm and would surely not give

birth to fruitful insights and a better understanding of the life of others.

25

Lützhöft (2004) also mentions the insider bias and how preconceptions about the world or a

particular domain shape the questions we do or do not ask and how we perceive the world around

us. This could perhaps be seen as a kind of professional vision (Goodwin, 1994), which might unlock

doors that would remain closed to a person lacking knowledge of the particular domain or trade –

who might fail to unpack the practice and knowledge of the practitioners studied (Lützhöft et al.

2009). At same time might an insider overlook to ask or feel awkward about asking perhaps

seemingly trivial questions; even though relevant information can be found (Lützhöft, 2004). Overall,

it is important to be aware of the both the positive and negative aspects of being an outsider /

insider and try utilize this knowledge the best possible way.

26

5 Analysis The analysis chapter is divided into two main parts, each section presenting the analysis and

discussion from the corresponding data analysis. 5.1 to 5.4 will cover the first and more naturalistic

analysis 1 and 5.5 and 5.6 will present and discuss analysis 2 that use the barriers and incentives

framework presented by Barach and Small (2000).

5.1 Analysis 1 The collected material was transcribed and is divided between designated persons and officers that

managed reporting on board. Statements deemed relevant were highlighted and subcategorized into

larger themes. The highlighted themes were:

Near-miss reporting

o Why is reporting important?

o Perceived barriers

o Means to increase reporting

Feedback and communication

Anonymity and confidentiality

INSJÖ

External parties that might affect reporting

Each theme has its own subthemes or subcategories that will be presented below.

5.2 Reporting officers onboard The material gathered from the officers is presented within the subcategories used during the

analysis described in 4.3.1. Each statement has been numbered to correspond to the representing

officer. This makes it easier for the reader to get a sense of which representative that is connected to

a specific statement without compromising the anonymity of the respondents. The numbering also

makes it possible to know whether a representative has more than one statement in a subcategory.

5.2.1 Near-miss reporting - Why is near-miss reporting important?

Near-misses is a familiar notion for all interviewees and is regarded an integral component in the

safety work on board. All ships have some type of goal regarding near-miss reporting. All officers

mention near-miss goals dictated from the office, three of the respondents whom work on tanker

ships also mention the similar goals from companies in the oil industry. The oil companies have their

own demands and regulations regarding near-miss reporting and are able to deny certain ships work

opportunities if these demands are not met properly.

A recurring notion from all four of the interviewed officers is that the reports, in particular near-

misses, should contain some degree of relevance, be useful and not trivial.

1. I am sure incidents occur that are not experienced as such and don’t get reported properly

for that reason. We have several sailors and deckhands that have worked onboard ships for

like 25-30 years. There is working tasks that I think are a bit unsafe, working tasks these guys

have encountered during their years at sea and therefore deem harmless. Of course they

27

won’t report this.

2. All reports must be relevant. You report all that’s relevant.

4. I mean, you don’t report if it’s not important. The report should be of serious matters and

useful. You always have to draw a line around what’s interesting to report. We often discuss

this onboard, especially when someone wants to make a report.

4. The written account should be useful. I mean, some sailors write about stuff that is nothing

to write about. Some might write too little, what do I know.

3. The work at sea is risk taking all the time. Something might happen that’s a bit out of the

ordinary, but you still feel it’s nothing to report.

Three out of four mention the learning aspect or lessons learned to be an important incentive to

near-miss reporting and two specifically mention that knowledge about specific events must be

spread to ‘sister boats’ with similar architecture.

2. There is an expression, “lessons to be learned” that emphasizes the importance of

reporting. The more reports sent in the less near-misses, you learn from your mistakes.

4. The system is made to share and teach others of your own mistakes

3. It’s quite important that we report. If you encounter strange things onboard that you aren’t

used to it’s a good thing to report. There might be others who need to know.

4. I mean, if you got a whole series of similar ships and there is something wrong with one of

them or if a near-miss occurs, then it’s pretty important to report this to prevent the same on

the other ships. It’s better to learn from each other.

3. If the other boats have similar systems and if you by happenstance discover ”This has

happened, this must be known” and you don’t report it, then the other boats can be

endangered as well and you haven’t shared your experiences with others.

5.2.2 Near-miss reporting - Perceived Barriers

One barrier that all officers mention is related to the near-miss reporting and the perceived

relevance and where to draw the line between a near-miss and ordinary events.

2. As I said before, near-miss reporting is pretty new in an ancient trade. I think it will take

some time before all people have the ability to discern a near-miss.

2. All reports must be relevant. You report all that’s relevant.

3. The work at sea is risk taking all the time. Something might happen that’s a bit out of the

ordinary, but you still feel it’s nothing to report. Often you feel, ”Well, that worked alright”.

4. Well, people hesitate to report, it happens. But it’s not a big deal, generally everything

works fine, it’s the ordinary life. If something happens, it’s often something that stands out,

minor stuff happens every day right?

28

4. There are always dangers onboard, especially during winter. And if you do something

wrong the risk for accidents is always there. It’s a special work place and really hard to know

where to draw the line. Sailors have always been practical; we are not a writing people or

office personnel.

1. You have to draw a line all the time. Sometime there is mooring and unmooring in strong

wind where risks for incidents are heightened. But still, in this profession, where do you draw

the line? It’s something no one has explained to me through the years, that line you have to

draw for yourself.

Another issue that might affect reporting is differences in the background of personnel onboard. The

educational and cultural background as well as age differs greatly between crew members according

to three officers. This might entail barriers such as different perspectives on the importance of

reporting, unfamiliarity with paperwork, feeling ashamed of reporting self-made errors or a

reluctance to forward near-misses or deviances to officers onboard.

3. There might be barriers against reporting for some. I have had the feeling that some

nationalities don’t want to admit when they done anything wrong. They’ll deny it.

4. Yeah, the crew members talk to the officers when they want to report something. There

are many people from other nationalities in this crew, just one or two Swedes. They aren’t

fond of writing, I don’t know why, it’s not like someone impedes them from reporting.

4. It’s very rare that someone in the crew wants to report something, I don’t know if they

might be a bit lazy in themselves or if they are scared of reporting or what it is. It’s always we

officers that have to report stuff, even if you try to encourage the crew. It happens really

rarely, I don’t think a deckhand has come forward to me and said that he wants to do a

report, ever. I don’t think it’s because we are officers, they aren’t used to writing or reporting.

4. Some people may feel that the report verifies that they are stupid, but you must not see it

that way.

1. The case might be that the individual worker doesn’t experience certain events as

incidents. He may have 25 years of service on the same boat and experienced these things

many times over during this time. He may feel that these things aren’t big deals, but someone

from outside may think this an obvious thing to report.

1. Well, the background might differ. It depends on the education you got with you from your

career. There are people that start out just after finishing primary school, or those that not

finished it before they started working onboard. They have little theoretical knowledge and

might not see this safety work the same way as those that read about it before they went out

to sea. Everyone has different qualifications. When talking about the merchant fleet the crew

composition is very varied. You got deckhands that might have scarce theoretical knowledge,

cooks, engineers and motormen. All with different educations or in some cases no education

at all. We also got the officers that have to go through merchant marine academy before

they even can start to sail. This mix might be one reason why not everything gets reported.

29

5.2.3 Near-miss reporting - Means to increase reporting

The interviewees mention that the changing focus on reporting as well as the improved

(navigational) safety is attributed mainly to the external parties and legislation. Two interviewees

mention that increasing demands from oil companies have helped moving the focus away from only

accidents and incidents to an approach involving proactive safety work and near-miss reporting.

2. The oil companies and the Swedish Maritime Administration have high demands on safety

and reporting.

1. The oil companies have their own inspections onboard that take place on a yearly basis. If I

remember correctly, they have conducted office inspections as well these last two years. In

relation to this I feel follow-ups, safety work and feedback have gotten much better.

1. Whatever your feelings towards the oil companies, one can conclude that the navigational

safety is much better on tankers compared to other ships. So they have brought with them

good changes to safety work.

5.2.4 Feedback and communication

According to all officers, the SMS meetings (meetings regarding safety onboard employed monthly in

most cases) are an important opportunity where feedback is given to the whole ship crew and the

members of the crew are able to voice issues and propose improvements that will be forwarded to

the DP.

3. Yeah, we got these SMS meetings monthly. We follow an agenda and talk about the safety

work. If something’s happened, you mention it. It’s the way things work, there’s no secrets

and no hypocrisy.

2. We continuously work to improve safety. We got these reports and the meetings onboard

to accomplish this.

1. We have a safety meeting once a month. On this meeting we talk about all SAFIR reports

the ship has sent since the last time. And if we have gotten info about near-misses from the

office regarding other ships the whole crew is informed during this meeting. Every crew

member has an opportunity to share experiences if something has happened. If they bring

something up, we have to report it.

4. We do a risk assessment each time there’s a safety meeting. Then you bring up everything

you think is serious or risky.

One positive feature with the reporting system that was mentioned by three of four officers is that

every report sent in must be processed by the DP or someone else in the office. The report keeps its

‘open’ status and remains in the system until someone deals with it. This keeps unsolved reports

visible in the system with lesser risk of being forgotten or neglected.

1. The system that handles SAFIR reporting shows a report as an open case until DP or

someone else at the office have made a decision. When we get a reply and everything is in

order it’ll close. A SAFIR can’t be ignored without proper steps taken.

30

4. Yes, when we send in a report it is open until the reported matter is solved. The company

and we onboard support each other. We do the required changes or propose improvements

to be made. When our measures get approved DP writes some notes about this and then

closes the case.

3. Each report gets a reply, if no reply comes back it’s an error in itself.

Two out of four officers mention that the DP role is most important when events that are more

serious occur. Furthermore, three out of four sees DP as the overarching safety coordinator and the

natural link between the ship and the company office.

3. DP is most important when bad things happen. When you report “the usual” you contact

the office people that work with safety related stuff.

1. Well, if your report concerns more practical stuff it’ll go to the company. If something

serious happens I’ll try to contact the DP personally or see to that he’ll be contacted directly.

3. DP should be available twenty four seven, seven days a week, if something happens. He

should assist us onboard concerning reports, feedback and whatnot.

2. DP handles the communication and information flow between office and ship. He’s the link

between the two.

5.2.5 Anonymity and confidentiality

Two out of four officers mention that anonymity is important to hinder pointing out individual crew

members when reporting self-made errors or slips.

1. It’s important that anonymity is provided to prevent someone to feel indicated. The whole

system would break down.

2. Fear of consequences shouldn’t prevent one sailor to write a relevant report.

In contrast to the two statements above, the other two officers do not feel that anonymity

necessarily is that important when reporting. Both officers mention that a seafarer should be able to

stand his or her ground and be responsible for his or her own reports without hiding his or her

identity, even though anonymity could be convenient in some cases.

3. I don’t like the use of anonymity. If you got something to say, you should stand your ground

and be honest. To say “this is what I think” instead of sneaking behind someone’s back. But

I’m sure there’s times where it might come in handy as well, because of the situation

onboard. There’ve always existed officers that are not fit for the job and idiots that made one

mess after another. Then it might be good that someone that ordinarily wouldn’t dare report

could give others a hint of the situation. There are both pros and cons.

4. I don’t think anonymity is necessary, if you want do a report or if you want improve

something, then you should stand for it. I mean, it’s nothing you just make up. There are no

lies and nothing unpleasant to sign your name for. So I don’t think it matters, I mean if I want

to report a near-accident for example I do it openly, it’s not like to badmouth someone.

31

5.2.6 INSJÖ

The officers mention INSJÖ only sporadically when asked about external parties that affect near-miss

reporting. The same officers talk about getting feedback from the database directly to the company’s

ships and that events from other companies’ database entries sometimes are discussed during safety

meetings.

5.2.7 External parties that might affect reporting

Only one officer emphasizes the issue of external influence. The interviewee brings up the two sides

of oil companies being so influent on the trade, at the same time as they have boosted reporting

they may also function as a brake when it comes to further improvement. The officer suggests some

kind of central instance that would function as a link between the shipping company and the oil

company. This central instance could also help change the focus to the learning aspects of reporting

rather than the potential negative consequences that might come with full transparency.

1. The oil companies often have higher demands than IMO on tankers. The oil companies

claims that it is voluntary to follow their instructions, they don’t have any legislation in their

back.. (..)Yeah I think barriers exist, I think it’s primarily the company that hesitates to

forward some things. The oil company inspectors are not very open, they are open to the

reporting system, but if something happens on a ship they might take this very serious and

not green light the ship. As they improve safety they also function as a braking block for

reporting. I feel that they sit on two different chairs at the same time. I don’t know how

things are in aviation now, but when I worked in the navy you had anonymity. The ones who

investigated accidents didn’t need to know the specific ship to improve safety. This could the

oil companies learn from, you could introduce legislation and recommendations that improve

safety without them knowing what specific ship did what. Today they know the specific ship,

which can be seen as something negative.

5.3 DP personnel Each statement has been numbered to correspond to the representing officer. This makes it easier

for the reader to get a sense of which representative that is connected to a specific statement

without compromising the anonymity of the respondents. The numbering also makes it possible to

know whether a representative has more than one statement in a subcategory.

5.3.1 Near-misses - Why is near-miss reporting important?

All seven interviewees stress that the learning aspect of near-miss reporting is of great importance to

be able to work proactively with accident prevention. Five out of seven DP personnel also mention

the ability to proactive work as the cornerstone of near-miss reporting.

7. Many near-misses will occur before the real event happens so to speak. If it’s possible to

process this when it’s still an almost-event it’s also easier to avoid the accidents.

7. Most of our work is focused on what happens on the ships, that we’ll be able to avoid

similar events or hinder that something happens at all. So yes, near-misses are really

important.

32

7. Sure, that’s why you don’t want just incidents, you also want to know when it almost

happened something so you can prevent incidents proactively. That’s why near-miss

reporting is important.

4. I think that near-misses are the most important actually, that’s where you really can

prevent bad things from happening.

4. Reporting near-misses are an important way to learn from each other.

6. Reporting near-misses is paramount to us, it's then you can prevent real accidents so to

speak.

6. The most important is to prevent things from happening again and to learn from past

mistakes.

2. Events that other companies don’t learn anything from aren’t sent forward to INSJÖ. There

has to exist a general interest.

5. The reporting is a really important way to spread information between our ships. Through

the system you can share important safety related info among all company ships.

3. Near-miss reporting is great when the right things get reported. Then you can work

proactively and prevent before something happens for real.

3. The important part of reporting is to learn from and teach others.

1. I want to emphasize how important this form of reporting is to proactive work.

1. The most important thing with reporting near-misses, incidents and accidents is to prevent

that similar events happens on another ship and that you learn something from each

accident, incident or near-miss.

Two out of seven stress that knowledge from near-misses are only distributed within their own

shipping company and that knowledge from other shipping companies is not used, one mentions

that it would be hard to manage information from the whole domain capacity-wise.

5. The reporting is a really important way to spread information between our ships. Through

the system you can share important safety related info to all company ships. It's also

important to make all ships highlight the information being sent out. The safety info is only

shared within the company.

1. External feedback isn't that important. It's much more important with feedback within our

own company. I don't like the idea of "the whole fleet of Sweden". It would be too much to

process other companies’ reports as in INSJÖ. I don't think it would help making our safety

culture better, it might be more important to larger shipping companies.

1. We use INSJÖ to report in but not as a tool to send out information to our own ships. We

use our own database to that end.

33

Two other DPs emphasize information sharing between companies and the importance to pool

information that relates to ships of the same type, sister boats, regardless of the company.

7. It all comes down to what has happened, if the event for example is about equipment that

only exists on one or a few ships then it might not be as interesting to others. But if there's

something that's common on all ships it must be shared with everyone. That's a consideration

you have to make. I don't always use the feedback I get from INSJÖ and send it back to the

ships, it depends on the matter at hand. It differs from case to case.

6. Mm, we also learn from each other within the company. Is there an incident on one ship

and there's a sister ship or a similar ship then we send out this information to everyone and

says: "This has happened at ship number one and we want you to take a look on this to

ensure that it doesn't happen to you as well".

One person stresses that near-miss reporting is a good indication of ‘safety awareness’, in the sense

of that a large amount near-misses reported statistically proves that the company has a good and

sound awareness of safety.

1. A year ago we had quite few near-misses. We ran a campaign that emphasized how

important it is to report near-misses. Both as proactive measures but also to show that the

crew have a safety awareness, then you can prove to your clients through good near-miss

reporting that thoughts of safety exists.

5.3.2 Near-misses - Perceived Barriers

All DPs mention barriers to reporting. The most prevalent type of barrier, that all except for one

respondent mention, is the noticeable variation in reporting among their own fleet. Certain ships or

individuals on ships are noticeably better at near-miss reporting.

7. We have explained to the crew that it's really important that they report. You can see a rise

in reports for a while. However, it's very individual what you define as: "This was nothing

special", or "This could have lead to something dangerous, I'll report this". Some are better

than others at reporting, it's very individual.

4. Yes, there're differences in reporting, but not between ships but rather between different

persons.

6. Some ships are very good at writing reports and do this frequently, they are good at

coming up with solutions as well. This is the way you want it in a perfect system, something's

happened, you acknowledge it, you write a report and then solve the situation before you

send in the report and adds the solution in the report itself. It doesn't work that way, there

only a few ships that's on this level.

2. There are large differences between different ships. One ship has reported almost one

hundred near-misses this year, whereas the others are close to the yearly goal. There's some

over-representation of near-misses on deck. Deck officers report a bit more often than chief

engineers and engineers, this might have to do with a better habit of reporting on deck.

5. There are better and there are worse ships when it comes to reporting.

34

3. Some individuals contribute to a better rate of reporting on some ships.

Four out of seven DPs also highlight the crew members’ background differences as a potential

barrier. Differences in age, culture and education are mentioned as factors that affect near-miss

reporting as well as trust between the office and the boat.

2. Deck officers report a bit more often than chief engineers and engineers; this might have to

do with a better habit of reporting on deck.

1. Some cultural differences exist between for example Swedes and crews with other

nationalities, there you could do a little extra work to enhance reporting.

7. I want to believe that reporting are easier for the newly educated officers. I hope the

importance of near-miss reporting is included in the education. I know that the schools have

access to INSJÖ and I hope that the ones that study now have this with them in their

backpack. The older officers may not see it the same way; they haven't seen the profit that

comes with reporting.

7. I think that newly graduated officers are more prepared to do things like risk assessments

and reporting of near-misses. But there might be two sides, this new generation has learnt

new ways, but they lack experience. The experience onboard is worth a lot as well, and is

something that takes many years to build up. You can't say that one generation is better than

another, but you have to acknowledge the differences.

6. You mentioned education, I should say that education is really important. You gain a

broader insight which I think lead to better reporting.

Four out of seven mention that understanding the motivation behind reporting can be improved

whereas three mention specifically that unfamiliarity with paper work, or in extension lack of

computer skills, might be an obstacle.

3. The near-miss reporting is going well, but you can always become better. You could try to

make the crew more aware of near-miss reporting, by setting monthly goals on reporting for

example.

1. I feel content with the quantity of reported events. Nevertheless, to increase reporting even

further people must understand why it's important to report, you could for example arrange

events to brace the crew.

4. Reporting is relatively new thing that people aren't used to. It takes time before the crew

understands the need and use of reporting. But you more you inform about reporting the

better they become.

6. You mentioned education, I should say that education is really important. You gain a

broader insight which I think lead to better reporting.

1. The crew sometimes thinks it's tiresome to write down stuff in the report form.

5. We would not manage to process sent in reports that are not correctly written.

35

3. There are still some people onboard, older people, that never have used a computer.

There’ll not be as many reports from these people.

Another issue mentioned by three of the DPs as being problematic when reporting, is how and when

one should draw a line between ordinary work conditions and near-misses; questions arise regarding

relevance, usefulness and non-triviality.

7. It depends on what you define as a near-miss when something happens onboard. Many

events could be seen as ordinary ones. You have to take a step further and think: "Could this

lead to something?" I think it is very individual what you acknowledge as a near-miss.

6. You have to draw a line for what you send in. I usually think: "What's the value to know

that this has happened at another company's ship?"

3. Near-miss reporting is great when the right things get reported.

Three DPs also mention that some crew members might feel that they ridicule themselves when

reporting certain events.

5. It might be hard for some people to write down near-misses, you get the feeling that you

ridicule yourself.

4. You don't want to be ridiculed by your colleagues, that won't do. That's the way it works in

society.

7. I think it's hardest when the near-miss is self made, you don't want to shout out to

everyone that you've done something stupid.

5.3.3 Near-Misses - Means to increase reporting

The single most prevalent measure to increase reporting is mentioned by five out of seven

interviewees. They emphasize that the crew members have to be aware of the importance of near-

miss reporting; or in two cases more specifically the importance of a (no blame) culture for safety.

This DP describes the creation of a good safety culture as a continual process that stretches over

many years. Two mention that the topic of near-misses are one of the focuses during SMS-meeting

and that the near-miss awareness could be improved by education.

1. We ran a campaign that emphasized how important it is to report near-misses.

3. You have to get everybody to know that near-misses are important. We measure statistics

on reporting for example. We set monthly goals to get people to be more aware of the

importance of reporting.

7. We have a no blame culture, no one gets blamed if you do something the wrong way, if it's

not criminal of course. But you don’t blame someone that makes an ordinary mistake. This

indicates that people really speak up when something's happened. We don't get mad if

something happens, it's always better to let us know. We continuously bring this issue to the

officers when we meet them. You have to explain what is done with the information that's

36

sent in and make them understand that the more reported near-misses the less risk of really

large events.

6. Well, as safety culture is made from several measures in the daily work process. It'll take

many years to build a good safety culture, and there are a lot of aspects and parameters that

you have take account of. It's nothing you can build in a year. But there's always room for

improvement. You mentioned education, I should say that education is really important. You

gain a broader insight which I think lead to better reporting. INSJÖ or this kind of reporting as

a whole is a continuous process to make more and more people aware of the purpose of the

system. This view is far from obvious for the crew onboard.

4. Reporting is a relatively new thing that people aren't used to. It takes time before the crew

understands the need and use of reporting. But you more you inform about reporting the

better they become.

Two interviewees, as mentioned in 5.3.2 above, note that newly educated master mariners probably

have more knowledge regarding maritime safety in ‘their backpacks’ and are therefore probably

more naturally inclined to reporting near-misses.

5.3.4 Feedback and communication

Most DPs assert that firsthand feedback regarding reports from the DP goes to the reporting officer

onboard the ship, even though several DPs also mention that information deemed important such as

specific near-misses goes out to internally by e-mail to the whole company fleet. One DP mentions

that feedback from INSJÖ is included in the DP’s reply to specific ships if the report from these ships

were forwarded to INSJÖ in the first place.

4. Yes, if there is a case that's of general interest I'll forward this feedback from INSJÖ

7. I don't always use the feedback I get from INSJÖ and send it back to the ships, it depends

on the matters at hand. It differs from case to case.

6. The feedback is included in the report that's e-mailed back to the ship. It's not only my

feedback but the information from INSJÖ as well. I don't think you always go through the

attached feedback from INSJÖ onboard. Primarily you look at the feedback directed to your

reported case specifically. You make sure that the preventive measures are taken care of,

that's what you prioritize. But I'm sure they look and ponder the INSJÖ feedback as well. I

don't work actively with that feedback that is sent from INSJÖ centrally, it's the specific ship's

job.

The two DPs that represent companies outside the INSJÖ collaboration describe more of a closed

loop of reporting. Near-misses regarded as relevant to notify about are only sent to the company

fleet. In one case, information is actually sent in to INSJÖ, even though the feedback from INSJÖ

never reaches back to the company’s own ships.

5. The reporting is a really important way to spread information between our ships. Through

the system you can share important safety related info to all company ships. It's also

important to make all ships highlight the information being sent out. The safety info is only

37

shared within the company.

1. I use INSJÖ to report in but not as a tool to send out information the company ships.

Two DPs mention that the feedback that goes from the DP directly to crew members (and back), is

mostly given through SMS-meetings. Three of seven DPs mention that the crew and officers are

encouraged to send in suggestions of improvement and not only report near-misses.

Three DPs mention that the DP functions as a direct line between the crew and the management,

though two of three have never encountered a crew member that use this method to contact

management instead of contacting the officer onboard.

1. It’s almost always the officer in charge that report. Crew members can contact me directly,

even though I can’t recall this ever to happen.

6. (..)if you look at the rest of the crew you don’t see that kind of contact with me actively,

most of the time I talk to the officers onboard.

7. Yes, it’s happened. It has been nothing serious, but sure, I’ve been contacted by crew

members directly.

5.3.5 Anonymity and confidentiality

Four out of Seven DPs claim that their company has a no blame culture.

4. I think it's important not to create scapegoats. It's not what you're after.

7. We got a no blame culture, no one's get blamed if you do something the wrong way, if it's

not criminal of course

1. We often talk with the crew about the importance of a no blame culture, but it's hard to

know how every man on deck views the matter.

6. If you want anonymity you should have it. We have a no blame culture here in the

company. If you find that you or a colleague doing something wrong, no one should be

blamed for notifying this, we are just grateful that you bring it up. Then fix the situation

instead, that's the way we work.

Four of seven DPs mention that anonymity is important on an individual level, though one mentions

that a low number of anonymous reports are a good indication of a good and transparent safety

culture.

6. I think it's important with anonymity, on the other hand there's not a great many

anonymous reports that get written. It's just a few on a yearly basis. I feel this is a sign of

transparency in the company. You can talk about safety, regardless. This could be a sign of a

good safety culture. I believe that a good safety culture is more of importance than

anonymity itself. If we didn't have a good safety culture there would be more people who

wanted anonymity, for sure.

38

One person mentions that anonymity and confidentiality is more important on a ship level than on

an individual level, and the fact that INSJÖ is closed to the public prevent denigration of shipping

companies in the press. Another DP, that do not use INSJÖ, comments that the maritime domain in

Sweden as a whole is rather small; which makes anonymous events in INSJÖ less anonymous at

second glance.

7. Anonymity is very important. Else I think we would lose reports. When you have a reporting

system the size of INSJÖ, there are many that can go in and have access to this information.

You don't want to reveal yourself and say: "this has happened on my ship and this has

happened on my ship". There's a risk that this kind of statistics is used to list all the problems

one company has had, statistics that would be printed in a newspaper.

1. The trade is not very large and you can sometimes deduce which company what despite

anonymity in systems like INSJÖ.

5.3.6 INSJÖ

All five DPs that report to INSJÖ mention that not every report is sent to the database. Four of these

DPs mention that reports they deem to have no value to others are not sent in to INSJÖ; whereas the

remaining DP mentions that he reports all of the incoming near-misses regardless. One DP only

sends ‘good‘ near-misses to the database.

7. All that's of value to others, things that could have ended really badly. That kind of things I

report to INSJÖ.

4. There's events of a more private nature, that's not for all eyes to see, not even within the

company. These kinds of reports we don't send forward to INSJÖ.

6. I don't send in all reports to INSJÖ. I decode like seven out of ten reports that I send to

INSJÖ. And if you ask: “why not ten out of ten?” Sometime the report has no value for others,

neither INSJÖ nor the reporter; in these cases I choose not to send the report forward.

3. All reports don’t get sent to INSJÖ. There are superfluous reports that just bring up things

like spelling errors. But all accidents and near-misses are sent to INSJÖ.

2. I choose to report to INSJÖ if there's a good near-miss. If it's too specific to one ship, to the

company or if there's no general interest, I don't forward to INSJÖ

Three out of five interviewees mention that they use or would like to use the feedback from INSJÖ

for risk assessment purposes in the future.

3. We haven't directly used the feedback from INSJÖ, it's mostly been used to show others

what can happen. No “aha” experiences. The statistics can be good for risk assessment

purposes, to know if it's happened earlier and so on.

Two interviewees mention that there exists a wide gap in quality between reports sent in to INSJÖ.

One of the DPs sees this as a crippling factor that limits the use of the database, whereas the other

DP sees a potential to learn something from every report.

39

3. One problem with INSJÖ are large differences in quality among the reports, some of them

tells you nothing of value.

4. The reports from INSJÖ come in varied quality, but it's still interesting to go through them.

You learn from everything.

5.3.7 External parties that might affect reporting

Two DPs mention the demands of certain quantities of near-misses reported each month to reach

the demands of the oil companies and of the inspections from the same companies on yearly basis.

The first participant also brings up an example of a near-miss case that he depicts as debatable.

5. We work with the oil companies, they want three to five near-misses a month.

3. The oil companies want at least one near-miss a month to show that the company has the

correct thinking

5. I’ve seen that it's not clean enough onboard. I reported this as a near-miss. You might view

this as: "ok, this might not really be a near-miss", but we feel that it is.

One DP mentions the double-edged role of the oil business. Another DP mentions that oil companies

are not allowed to use a report against a company even though problems of this nature have

occurred in the past.

5. Oil companies aren’t allowed to use the reports against the company, this rule is better

followed nowadays, but it hasn't always been so. Earlier the shipping companies didn't dare

bring some things up because of the risk of "decapitation".

3. All parties want the shipping companies to be proactive. The clients are always positive to

reporting until the report is sent in, you get the feeling that mostly negative critique come

back when it's sent in. It would be good with more support.

5.4 Discussion - analysis 1 I will here discuss material from analysis 1 that is mentioned by several respondents or notions that

are emphasized by one or very few respondents but deemed relevant on their own. The common

denominator is that I see these notions as relevant for this study. A discussion of analysis 2 will follow

in 5.6.

5.4.1 Benefits and barriers to near-miss reporting

DPs and Officers almost unanimously highlight the proactive power of near-miss reporting and that

lessons to learn from this kind of events are of importance. There are some notable differences

between DPs and officers that belong to the INSJÖ cooperation and those who do not. Companies

that are part of the INSJÖ cooperation mention advantages of having access to a nationwide

database for incidents and near-misses to a large extent whereas the companies that are not a part

of INSJÖ seem more content to spread the lessons learned within the boundaries of their own

company. This is relevant in regards to the tradeoffs mentioned by Barach and Small (2000), a

nationwide database might contain less relevant reports that lack in detail. However, rare

occurrences and ‘unique’ events, that might transpire due to a certain setup of factors, are more

40

likely to be found in a larger accumulation of reports. A local database might therefore seem more

efficient in the day-to-day work, as a company’s own reports probably make more sense to use in a

local context. Still, using a local reporting system (the company’s own reporting system) only might

be less effective as a proactive measure to learn about, and in a worst-case scenario hinder, rare but

maybe more disastrous accidents.

There might also be a division of philanthropic values (Barach & Small, 2000). Some companies

emphasize the value of sharing information (and learn lessons from other companies) within the

whole maritime domain whereas others find it more valuable to inform and spread knowledge

among sister ships and their own company ships. The importance of reporting being therapeutic

(having learning value) seems common among all companies and respondents, the learning factor is

held high by practically all the respondents.

Evans et al. (2004) discuss the lack of organizational feedback that was identified as a major barrier

for doctors and nurses. Lack of feedback is not mentioned as a particular barrier to near-miss

reporting in the maritime domain by the respondents. The regulative function of the DP and the

nature of the reporting system create a stable feedback stream between DP and officers onboard, as

reporting matters must be dealt with and closed through the SAFIR system. The present study does

not however answer the question whether crew members in general feel that they get adequate

feedback from officers or the DP. Some officers mention that crew members of certain nationalities

report less or not at all or that differences in crew members’ background might cloud the larger

safety picture for some.

Van der Schaaf & Kanse (2004) emphasize differences in perceived reasons for not reporting

between management and operators. When presented with the results in van der Schaaf and Kanse’s

(2004) study most management and safety personnel were surprised that the major reasons for not

reporting did not involve fear of reporting or avoiding shame. There are some differences in

perceived barriers between DPs and officers in this study, but also an agreement of factors that

affect reporting in the maritime domain negatively. Three out of seven DPs mention that uneasiness

and fear of shame might be a barrier, whereas only one officer out of four mentions very briefly that

this might be a problem. A noticeable difference in this study compared to van der Schaaf & Kanse’s

(2004) is that the respondent groups, DPs and officers, might be perceived as authority figures or

management, from other crew members point of view. Differences in opinions perceived in this

study are between the two respondent groups, DPs and officers. The view of the rest of the crew

remains unknown. The IMO guidelines on near-miss reporting (2008) also mention fear of shame as a

barrier for near-miss reporting and as one major reason for the need of a just culture. With lack of

input from the crew members, it is difficult to know how prevalent this barrier is within the

companies in this study

Other noticeable potential differences are presented in the anonymity paragraph below. One very

prevalent notion that is mentioned by all DPs and several officers is that reported events must be

relevant and not trivial. At the same time, where do you draw the line between ordinary working

condition seafarers are used to and near-misses? One of the dangers mentioned with near-misses is

their ’trivial’ nature. Triviality might create resonance in a system’s stability and become hazardous in

orchestra with other factors. The problem of judging the seriousness of an event is noticeable in the

41

literature and something seemingly common in various domains such as family medicine (Elder et al.

2007), the chemical process industry (van der Schaaf & Kanse, 2004). Barach & Small (2000) list

skepticism of near-miss reporting as one cultural barrier that is prevalent in aviation, petrochemical

processing, NASA and the nuclear power industry.

Several DPs mention that a good way to achieve an increased near-miss reporting is to enhance the

awareness of reporting near-misses. This might seem simple, but could be complicated in reality, and

should be addressed and treated seriously. To alter one’s view both an understanding of the matter

at hand and effective means to bring about change is needed. One danger is that the ‘importance of

near-miss reporting’ is rehearsed but the view of the matter remains the same and reporting is

regarded mostly as extra paper work. As Sanne (2008) concludes, to make reporting systems work

effectively, a focus should be to give employees ownership of their own reporting system and the

knowledge how and why to use it. Sanne (2008) mentions the storytelling scheme as the

occupational norm in the railroad domain. While storytelling is a way to address risk- and surely a

possible tool to unravel knowledge that might disappear in a more formal reporting scheme - this

perspective focuses on risk in a way that favors a more narrow and local perspective rather than a

wider one.

Another aspect that both DPs and officers mentioned during the interviews is the difference in

reporting between ships and that some officers are more used to reporting than others. This might

not come as a surprise in a domain with many generations of officers working onboard ships. There is

a large age span among members of the maritime domain and this could mean a variation of skills,

experience and knowledge. Both DPs and officers mention differences in background, such as

difference in crew education, as a factor and potential barrier that influence how safety is viewed

onboard.

5.4.2 Feedback and communication

The communication in reporting matters between ship and office mainly between the DP and the

officer in charge of reporting matters. Three DPs assert that the crew can contact them directly

without notifying either officers or office, though only one DP claims that use of this privilege has

occurred during his time in office.

The feedback process between DP and officer within the reporting system SAFIR seems to function

well according to most officers. According to all four officers and two DP feedback to other crew

members stems primary from safety meetings onboard.

SAFIR reporting is described as a built-in measure to get feedback on all reports sent in. Most near-

miss reports are forwarded by a DP to INSJÖ in the cases where the company is a participant in the

INSJÖ cooperation. One DP mentions that all near-miss reports are sent in whereas another only

sends in the ‘good ones’. This might lead to the same problems with triviality and where to draw a

line between what events are valuable to share with the whole community and which ones are not.

The feedback stream from INSJÖ back to the actual crew seems to differ between companies and

ships. One DP mentions that an e-mail with INSJÖ recommendations is automatically sent to the ship

where in other cases it is up to the DP in charge to choose whether to inform certain ships or not.

Johnson (2003) mentions the risk of bias towards certain groups in national systems, something that

can be perceived in this domain as well. Not every Swedish shipping company is part of the INSJÖ

42

cooperation and there is quite a difference in quantity of how many reports that are sent to INSJÖ as

well as how actively companies use the retrieved information in their daily work. A risk with national

systems that are difficult to examine further in this study is the possible discrepancies in mentioned

and actual use of national reporting systems as a tool for accident prevention. Johnson (2003)

mentions similar risks when he describes how national reporting systems might develop into

’grandiose initiatives’ that fulfill the ambition of its proponents rather than addressing safety issues.

5.4.3 Anonymity and external parties

When talking about anonymity and confidentiality there are some differences in perspectives

between DPs and officers. Most DPs talk about anonymity and confidentiality in positive terms and

as means to decrease blame and improve reporting at large, though one DP mentions that he is

feeling a bit divided regarding anonymity. He explains that he feels a bit of skepticism against

anonymity. He mention that it is important in the sense of providing anonymity. Still, the crew should

at the same time not feel the need of reporting anonymously if there is indeed a no blame culture

installed and a sound safety culture prevalent in the company. In contrast to this, two officers oppose

anonymity with rather strong words and emphasize that one should stand his ground and take

responsibility for what has happened without a layer of anonymity. The mentioned issues with

anonymity seem to concern responsibility rather than the paradox of anonymity (Johnson, 2003)

mentioned in 2.3.

When anonymity is highlighted as a topic during the interview, the respondents seem to focus on

anonymity on two levels, the crew level and the company level. The focus of anonymity mostly

concerns the crew level where it is mentioned as a necessary component in the reporting system.

One officer mentions anonymity as important on a company level as he is talking about external

parties, in this case the oil companies. He mentions that the oil industry affects the reporting in both

positive and negative ways, this due to their power to demand changes regarding safety and the

authority to deny companies working possibilities if certain standards are not met. This officer

suggests that there should be another layer of anonymity added towards these companies in a

central autonomous manner, a model not wholly unlike the INSJÖ system. This would regain the

lessons learned effects but dampen the negative focus on specific ships that might come from full

transparency towards the oil business. Two DPs mention monthly near-miss demands specifically,

whereas one further comment that reports have been used against companies earlier to ’decapitate’

the company. This potential negative external effect combined with a certain vagueness of the

nature of near-misses might create a problem of relevancy. Does strict demands of monthly near-

miss reports equal safety or sound safety awareness? Is there a risk of ‘finding’ debatable near-

misses out of necessity due to these requirements but still dismiss others, especially when personnel

might not share the safety view of the DP or certain officers, due to their perceived triviality?

The use of INSJÖ seems to vary from company to company. Three out of five DPs use the feedback to

INSJÖ for risk assessment purposes though two out of five officers also mention that there are

significant differences in quality between different reports in the INSJÖ database. This creates a

problem of not only getting people to report near-misses, but also how to present the reports in a

manner that is useful for the Swedish maritime domain.

43

5.5 Analysis 2 – Barriers and incentives to reporting The second analysis applies Barach and Small’s (2000) framework of barriers and incentives to near-

miss reporting, presented in table 1 in 2.3.5, on the gathered material. This analysis intends to

highlight the barriers and incentives identified in this study in comparison to barriers and incentives

found prevalent in other domains.

The four main categories used in this analysis concern barriers and incentives within the legal,

cultural, regulatory and financial area. Each main category contains subcategories – marked in bold -

that in turn includes notions related to the collected material from the study's informants. The

presented companies are de-identified and referred to as numbers. Company number one and five

do not use the INSJÖ system in their accident prevention work. Each respondent group has

statements regarding barriers and incentives on the individual and the organizational level as can be

seen in each group’s two separate columns. Subcategories within brackets are categories used in the

study of Barach and Small (2000) even though they have not been mentioned by the informants or

have been identified in this study. To get a better overview of each main category table 2 below has

been divided and is presented in four parts.

44

Table 2.1. The legal category of table 2 with both barriers and incentives to near-miss reporting

DPs: (1),2,3,4,(5),6,7 Officers onboard: (1), 2, 3, 4

Individual Organizational Individual Organizational

Legal

Barrier Fear of reprisals:

Fear of negative

consequence when

reporting (1, 4, 5, 7)

Lack of trust:

Supervision when

reporting (6)

Tension between office

and crew (7)

Sanctions:

Risk of negative

consequences from oil

companies when

reporting (3, 5)

Bad publicity:

The layer of anonymity in

INSJÖ might not be

enough to de-identify

companies due to the

small domain (1)

(Undermine trust)

(Fear of litigation)

(Costs)

Fear of reprisals:

Hard to maintain

anonymity when the

crew is small: (1)

Crew might omit

reporting, don’t want

to get involved: (3)

Crew from certain

cultures don’t like to

report self-made errors

( 3, 4)

Lack of trust:

Crew from certain

cultures don’t like to

report: (3,4)

(Fear of litigation)

( Costs)

(Sanctions)

(Undermine trust)

(Bad publicity)

Incentive Provide confidentiality

and immunity:

Must not create

scapegoats: (4, 5, 6),

Provide confidentiality

and immunity:

INSJÖ prevent negative

media coverage when

reporting (7)

(Provide confidentiality

and immunity)

(Provide confidentiality and

immunity)

The statements of the DPs and officers are categorized based on the mention of individual or organizational barriers /

incentives respectively, which create four dimensions of barriers / incentives

Barriers on the individual level found in the legal category mainly concern fear of reprisals and lack

of trust. The only incentive mentioned in answer to these barriers is to provide confidentiality and

immunity. Barriers on the organizational level are only identified in the DP group and concern

sanctions and bad publicity.

45

Table 2.2. The cultural category of table 2 with both barriers and incentives to near-miss reporting

DPs: (1),2,3,4,(5),6,7 Officers onboard: (1), 2, 3, 4

Individual Organizational Individual Organizational

Cultural

(values,

attitudes,

beliefs)

Barrier Dependent on

profession:

Where do you draw the

line during ordinary

working conditions: (1, 3,

7)

Different cultural

backgrounds: (1)

Different educational

backgrounds: (1, 3, 6 , 7)

Some crew members Lack

knowledge about DP role:

(3, 6),

Computer inexperience:

(3),

Crew (or officers) not

used to reporting: (4,5)

Skepticism:

Where do you draw the

line during ordinary

working conditions: (1, 3,

7)

Only reports “good

enough” near-misses: (2)

The report must be

useful: (7)

Extra work:

extra workload: (2)

(Code of silence)

(Fear of colleagues in

Trouble)

Dependent on

organization:

Feedback from INSJÖ

differs greatly in quality:

(3)

Bureaucratic:

Can not handle reports

that are not correctly

written: (5)

(Pathological)

(Generative cultures)

(Don’t want to know)

Dependent on

profession:

Where do you draw the

line during ordinary

working conditions:

(1,2,3, 4)

Differences in

background: (1)

Everyone does not

understand or value

near-miss reporting: (1,

2, 3)

Not used to paper

work: (4)

Skepticism:

Where do you draw the

line in the line during

ordinary working

conditions: (1, 2, 3, 4)

Everyone don’t

understand or value

near-miss reporting: (1,

2, 3)

Must be something

useful: (3, 4)

Extra work:

Not used to paper work

/ not a writing people:

(4)

(Code of silence)

(Fear of colleagues in

Trouble)

(Dependent or

organization)

(Pathological)

(Bureaucratic)

(Generative cultures)

(Don’t want to know)

46

Incentive Professional values:

philanthropic:

Share within the

company: (1, 4, 5, 6)

Share with similar ship

types: (6, 7)

Knowledge of general

interest must be shared

(and learned from

others): (2, 3, 4, 6, 7)

Integrity:

An understanding that

official reporting leads to

guaranteed feedback

from the DP (4)

Openness regarding

reporting regardless: (6),

Direct line to the DP: (7)

Educational:

Near-miss awareness /

emphasis: (1, 3, 4, 5, 6, 7),

Emphasis on near-miss

reporting on SMS

meetings and Officers'

conferences: (3,7)

Using cases from INSJÖ

and other scenarios in

educational purposes: (4,

6, 7)

Create a good safety

culture: (6)

Everyone should know

the role of the DP: (7)

(Cathartic)

Become a leader in

safety and quality; good

for business:

Safety has the highest

priority (6)

Professional values:

Philanthropic:

Efficient feedback to

company ship from

office (1, 2)

Knowledge of general

interest must be shared

(and learned from

others): (3, 4)

Share with similar ship

types: (4)

Integrity:

Reporting guarantee

feedback (1, 3, 4),

Straight forwardness

and honesty (is more

important than

anonymity): (1, 2, 3, 4),

SMS meetings are a

way for crew members

to voice their opinion

and 'clear the air': (1)

Open dialogue with the

crew: (3)

Educational:

Newly educated

officers have more

theoretical knowledge:

(4)

Cathartic:

SMS meetings are a

way for crew members

to voice their opinion

and 'clear the air':(1, 3,

4)

Lessons learned: (2)

Become a leader in safety

and quality; good for

business:

Striving to continually

enhance safety onboard:(2)

The statements of the DPs and officers are categorized based on the mention of individual or organizational barriers /

incentives respectively, which create four dimensions of barriers / incentives

Barriers found in the cultural category concern several subcategories. On the individual level, both

DPs and officers mention barriers dependent on profession, and in relation to skepticism and the

extra work needed to report. Both groups mention incentives relating to philanthropic values and

the importance of integrity and educational resources. Organizational barriers are few and are only

mentioned by the DPs.

47

Table 2.3. The regulatory category of table 2 with both barriers and incentives to near-miss reporting

DPs : (1),2,3,4,(5),6,7 Officers onboard: (1), 2, 3, 4

Individual Organizational Individual Organizational

Regulatory

Barrier Exposure to malpractice:

Fear of negative

consequence when

reporting (1, 4, 5, 7)

Potential censure:

Supervision when

reporting: (6)

(Premiums will go up)

(Investigation and

license suspension and

subsequent loss of

income)

It doesn’t apply to us,

we do our own internal

analysis process:

Using company’s own

databases reporting

systems only, not INSJÖ:

(1)

Not enough resources to

handle reports from the

whole Swedish maritime

domain: (1),

(They can’t understand

our problems anyway)

Exposure to

malpractice:

Hard to maintain

anonymity when the

crew is small: (1)

Crew might omit

reporting, don’t want

to get involved: (3)

Crew from certain

nationalities d not like

to report self-made

errors: ( 3, 4)

fear of ridicule: (4),

License suspension and

subsequent loss of

income:

Oil company demands

and regulations ( 1)

(Premiums will go up)

(Investigation and

potential censure)

(It doesn’t apply to us)

(We do our own internal

analysis process)

(They can’t understand our

problems anyway)

Incentive Prophylactic:

Strive for a No-blame

culture: (1, 7)

Anonymity important on

an individual level: (2)

Must not create

scapegoats: (4, 5,6)

Follow the rules:

An understanding that

offical reporting leads to

guaranteed feedback

from the DP: (4)

Emphasis on the ISM-

code: (4, 7)

Set goals in near-miss

reporting: (3)

Every near-miss is sent to

INSJÖ: (3)

Report as near-miss when

applicable: (5)

(Fear of censure) Prophylactic:

Anonymity is important

on the crew level,

names not mentioned

(1, 2)

It is human to err: (3)

Follow the rules:

SMS meetings is a way

for crew members to

voice their opinion and

'clear the air': (1, 3, 4)

Reporting guarantee

feedback (1, 3, 4)

The SMS manual is

integral to reporting:

(3)

The officers onboard

emphasize reporting:

(3)

(Fear of censure)

The statements of the DPs and officers are categorized based on the mention of individual or organizational barriers /

incentives respectively, which create four dimensions of barriers / incentives

48

Both groups highlight exposure to malpractice as a barrier on the individual level. Fear of censure as

a barrier is found in the DP group only whereas the risk of license suspension as a barrier is identified

solely in the officer group. Prophylactic no-blame directives as well as rule following are identified as

incentives on the individual level in both groups. The only barrier on the organizational level is

mentioned by a DP and is related to the view that the use INSJÖ is not important in his company.

Table 2.4. The regulatory category of table 2 with both barriers and incentives to near-miss reporting

DPs: (1),2,3,4,(5),6,7 Officers onboard: (1), 2, 3, 4

Individual Organizational Individual Organizational

Financial

Barrier (Loss of reputation)

(Loss of job)

(Extra work)

Potential loss of

revenue:

Risk of losing job

opportunities if demands

from oil companies are

not met (5)

(Wasted resources)

(Patient care contracts)

(Not cost effective)

(Loss of reputation)

(Loss of job)

(Extra work)

Potential loss of revenue:

Risk of losing job

opportunities if demands

from oil companies are not

met (1)

(Patient care contracts)

(Not cost effective)

(Wasted resources)

Incentive (Safety saves money) Improve reputation of

quality and safety:

Near-miss statistic / Oil

company inspection: (1,

2, 3, 5)

(Publicity relations)

(Safety saves money) Publicity relations:

The DP functions as a link

between company,

organizations and the

media: (1)

Improve reputation of

quality and safety:

The DP functions as a link

between company,

organizations and the

media: (1)

Near-miss statistic / Oil

company inspections: (1)

External organizations and

clients demands safety

focus: (2,4)

The statements of the DPs and officers are categorized based on the mention of individual or organizational barriers /

incentives respectively, which create four dimensions of barriers / incentives

The financial category only concern barriers and incentives on the organizational level. The risk of

losing revenue is identified as an barrier in both groups whereas found incentives are the

improvement of the company’s reputation of quality and safety publicity relations with the media

and other organizations.

Categories in this material that seemingly relate to each other using the general categorization from

Barach and Small (2000) are integrity and rule following. The best channels for the crew members to

49

be heard and voice concern without being ignored are through proper channels integrated in the

system. Following the rules by reporting through proper channels guarantee that feedback will be

received. The notions of straightforwardness and honesty are mentioned as incentives to reporting

by all the officers though the DP personnel do not mention this particular motivator at all. Questions

about where to draw the line between ordinary working conditions and near-miss events are

mentioned by both groups as a common barrier in the maritime domain. This barrier also seems to

relate to the notion of skepticism of reporting events that is regarded as trivial.

According to the Barach and Small’s (2000) categorization, anonymity, confidentiality and immunity

can be contemplated from various perspectives; this categorization highlights the legal point of view

as well as the regulatory and perhaps more local company perspective. The notion of no-blame

cultures seems to reside within each company, being prophylactic and not something that could be

said to necessary be a part of a larger legal category, it rather fits with the notion that ‘some degree

of immunity‘ (Barach & Small, 2000) is provided. Both officers and DP personnel mention barriers

related to this regulatory category. These barriers are mainly related to the fear of exposure of

malpractice from the crew. There is a seemingly small but relevant distinction between this barrier

and the related barrier of fear of reprisals that can be found in the legal category. The data suggests

that fear of shame and ridicule is more prevalent than fear of legal consequences in this case. It can

be hard to distinguish the two categories. Both groups mention ‘where to draw the line?’ as a

possible professional barrier to near-miss reporting. This suggest that many near-miss occurrences

are not viewed as something that lead to legal consequences to the same extent as they are viewed

as something trivial or ordinary. To omit reporting certain near-miss occurrences might relate more

to fear of embarrassment or not feeling the need to report than fear of legal consequence. The legal

category also include the subcategory ‘lack of trust’, a category that should, with above mentioned

relation between legal and regulatory fit well within the regulatory category as well; lack of trust

should be related to fear of exposure as much as fear of legal consequence.

The only other potential barrier found in the regulatory category is one of potential censure related

to a notion that crew members’ reports should be reviewed by an officer before forwarded to the

office. This notion might also be relevant in ‘lack of trust’ in the legal category.

The barriers connected to the subcategories dependent on profession and skepticism within the

larger cultural category are mentioned by both groups on several occasions; whereas the categories

professional values and educational in the same main category contain many incentives to near-miss

reporting. DP personnel focus more on educational incentives whereas both groups mention the

importance of philanthropic values. All DPs emphasize that knowledge of general interest to all

companies must be shared. The exception are the two representatives that do not use INSJÖ as part

of their safety work. They stress instead that it is critical to share relevant information within the own

company.

Two of the DPs mention that crew members or officers might not be used to report and two other

ones state that some members lack knowledge of the DP role. The latter might suggest that

knowledge about the DP is interwoven with the knowledge of using reporting systems. Three out of

four officers refer to a similar barrier where crew members do not fully understand or value near-

miss reporting.

50

That feedback is an incentive to near-miss reporting is apparent even though feedback do not claim

its own category. Still, it is hard to tell whether feedback should pose as its own category or if it is

more of a philanthropic and regulative function.

5.6 Discussion - analysis 2 Analysis 2, presented in table 2 in section 5.5 examines the gathered material through the lens of

Barach and Small’s (2000) general categorization of barriers and incentives to near-miss reporting.

This is done to compare findings in this study with identified categories from other domains that use

near-miss reporting.

The second analysis suggests that barriers related to categories concerning the fear of exposure are

more prevalent than the fear of legal consequences when reporting. This indicates that measures to

create trust and no-blame within the company and among the crew members are at least as

important as providing confidentiality and immunity by legislation.

The category ‘dependent on profession’, not surprisingly, is tightly connected to the domain or

profession under scrutiny. This category could be used to highlight and compare values, attitudes

and beliefs in other domains and organizational settings with those found in the Swedish maritime

domain, or specific parts of it. The categories ‘code of silence’ and ‘fear of other colleagues in

trouble’ are not mentioned directly by any of the respondents. They might still be linked to lack of

trust and the mentioned lack of reports from certain nationalities, as mentioned by the informants.

The fear of exposing colleagues is indirectly connected to the fear of one’s own exposure. Differences

in crew members’ background are highlighted as a barrier in both analyses. Even though this, for

most part, is mentioned in regard to the background of crew members and to some lesser extent to

officers. Education and an increased awareness of near-misses and proactive accident prevention are

presented as a remedy. This is hard to argue against. Questions relevant are how and what this

education should consist of. Are measures that aim to uniform background knowledge as well as to

‘educate’ certain crew members, so they will be able to distinguish what kind of events that are

important to report, the best way to go? Or is there a risk of losing knowledge if there is less diversity

in the ‘correct’ point of view? If the reporting system should be ‘owned’ by the crew members

themselves, which measures should then be used to pool together and highlight experience and

knowledge from different points of view without losing focus on how to use this information?

The financial and legal categories are not mentioned as much as the cultural category. There is no

doubt that there are more barriers that relate to these two categories, even though these barriers

and incentives for most part concern the companies’ relation to outside parties that might influence

reporting by making demands and affecting the company financially. It is the informants of tanker

companies, which work closely with the oil business mention this aspect to the greatest extent. The

only other statement in this category concerns how exposure to the media might pose a sensitive

barrier.

As in the first analysis, differences in philanthropic values (with whom the company shares their

reports and near-miss information with) between the DPs in INSJÖ and Non-INSJÖ become visible,

and seem to be the most relevant differences between these groups found in the collected data.

51

The division of individual and organizational barriers and incentives highlight the need to oversee the

entire reporting process; from steps to encourage the crew members to report, to explore the

process taking place within the office, as well as an examine the forwarding of selected near-misses

to the INSJÖ database. Categories that seem to relate to each other would also be harder to identify

if not for the categories used in this analysis; as the subcategories of ‘integrity’ and ‘rule following’, in

the sense that reports sent through the reporting system cannot go unanswered and must be given

proper feedback by the DP.

52

6 Discussion The discussion chapter is divided into four main parts. The chapter begins with a general discussion

in section 6.1 followed by a discussion in 6.2 that concerns the theoretical framework and the

discussion of concepts used in this thesis. The methodology section, 6.3, discusses the method used

and how the study was conducted. Section 6.4 suggests topics for future research.

6.1 General discussion The aim of this thesis was to study near-miss reporting and to identify barriers that might hinder or

limit near-miss reporting in the Swedish maritime domain. The statements from DPs and officers in

this study suggest that similar barriers to near-miss reporting can be found in the maritime domain

as in other domains, which use near-miss reporting systems, presented in the study.

Barriers in this study have been identified either by direct answers from respondents or deemed to

be found implicitly in what the respondents say. Two seemingly important threads that can be found

in both analyses are that of integrity and anonymity / confidentiality. These two threads are not

necessarily incompatible but might clash in some ways. A misuse of anonymity could stand in direct

opposition to integrity whereas integrity might not be enough if the fear of reprisal or exposure are

big enough (e.g. you risk being affected economically, risk being seen in a bad light or expose yourself

in some other negative way). On a local level, especially in smaller companies, this anonymity or

confidentiality risks being breached, by purpose or not, due to a small number of crew members and

vessels. The best way to remedy this problem might be hard to identify, but creating a genuine trust

in the company management and between crew members might play a key role.

Another barrier made salient in both analyses is the view on relevance and significance of near-

misses. All respondents have not mentioned the view of certain near-miss events as trivial as barriers

per se; but these statements have rather been interpreted as barriers by me. Literature from other

domains presents similar barriers, but does this type of skepticism need to be a barrier to reporting

by default? There might exist different types of ‘triviality’, on one hand there might be diffuse near-

misses that are hard to assess properly in a risk spectrum and on the other hand there might be near-

misses that really are trivial, that might be reported as a result of the growing focus on near-miss

statistics. This skepticism could perhaps also work as a professional vision filtering out certain trivial

events that actually are known within the profession as well as to rule out truly harmless events; this

all depends on how near-misses and their inherent nature is viewed. This will be discussed more

thoroughly in section 6.2.

The second analysis focuses more systematically on both barriers and incentives compared to the

first analysis and made it possible to compare DPs and officers. Legal and economic

barriers/incentives though, are scarcer than barriers/incentives in the cultural and regulatory

categories that hold more focus. Still the legal and economic aspects were not highlighted much in

the first analysis either, and is thus absent in both analyses.

One thing to have in mind is that the boundaries of this study are within Swedish maritime domain

and statements from other parts of the domain are not presented. This does not mean that the

creation of efficient accident prevention is an isolated national problem and not important for the

maritime domain everywhere. The use of reporting systems is local (within the company only) and

53

national (using INSJÖ), as can be seen in this study. The use of the third category, international

reporting systems, is not mentioned to a large degree in this study but may still be one important

component that should be available when learning from each other through reporting systems. The

task of creating efficient international reporting systems are not without complications, Johnson

(2003) describe common problems such as the difficulty of collating data efficiently when the parties

included use different terms, definitions and so forth; information that may be tailored in certain

ways to suit specific needs of each party. This might create a barrier, of transferring and sharing vital

information in-between cooperating parties that are difficult to bring down.

Potential problems with uniting all Swedish shipping companies under one voluntary reporting

system are apparent in this material. How should one undertake the measures to efficiently

cooperate on an international level when there are apparent obstacles to tackle on the national and

local level?

Is the issue of barriers to near-miss reporting in the maritime domain then an issue to handle

through legislation and recommendations on an international level, on a national level or primarily

on the company level? This study‘s focus has been on the respondents statements and not on the

organizations presented in 2.1, even though the interrelation between organizations working for

safety of course is of great importance in order to attain an effective and functional reporting

culture.

6.2 Theoretical Framework The concepts of near-miss, incident and accident are common within a large number of socio

technical domains, as have been presented in the literature overview in 2.2. Several of the DPs

interviewed mentioned ’Bird's triangle’ when discussing near-misses, and the accident-ratio model

seems to be an accepted motivator for why near-miss reporting is important. The proactive work is

also described as way to remove near-misses at the bottom of the triangle to prevent accidents

further up.

This stands somewhat in contrast to potential critique of the ratio triangles themselves, both the one

made by Bird and Loftus (1976) and the similar one depicted by Heinrich et al. (1980). Heinrich et al.

(1980) presents reasonable critique against the suggested generality and causal explanations of their

own accident-ratio model and the implications that causes of frequent incidents are the same as the

causes of severe injuries. Johnson (2003) further brings up potential problems of interpreting ratios

found in various domains; it is hard to know what the ratios actually tell, without proper context. If

you cannot generalize these ratios between domains or jobs, and the authors themselves refute their

own previous statements, it is difficult to know what kind of assumptions you could draw, at least

from the ratios alone. The data supporting Heinrich et al. (1980) was collected during the thirties if

not earlier; the first edition of their book was published in 1932 and has since then gone through four

more revisions. This fact could make it difficult to validate this data today. One can wonder if the

data behind the ratio of Heinrich et al (1980) is even meaningful to interpret in our constant changing

world where socio technical systems and views of human rights and lives have changed

tremendously since the time when the data was collected. Bird and Loftus’s (1976) accident-ratio is

based on 1 753 498 accidents reported from 21 industries, this overwhelming amount of reports

54

might give power to statistical analysis, but also create questions regarding reporting format,

reporting standards and reporting quality, that goes unanswered in this study.

When looking at systemic views of socio technical systems such as Perrow (1999) it might be hard to

maintain the underlying accident-ratio model of Heinrich et al. (1980) and Bird and Loftus (1970) as a

plausible explanation for the occurrence of accidents. Most proponents of a systemic viewpoint of

socio technical systems probably regard the model as too simplistic a model of ‘reality’. This does not

mean that the accident-ratio model are ‘wrong’ but rather that the complexity of socio technical

systems, their interactions and couplings, are factors that are left out from the model. If the focus

moves to a more complex model explaining why accidents happen, there should be careful

considerations of what this shift in focus means in regard to concepts such as accidents, incidents

and near-misses as well. Many proponents of a systemic viewpoint describe accidents and other

events as resonance within the system, as well as something that ‘emerge’ and consequently might

be hard to predict based on the state of every single component in the system (Dekker, 2010). This

systemic stance makes one integral key component of the accident-ratio model problematic - that of

‘cause and effect’, as no parts need necessarily to be broken for the system to brake (Dekker, 2010).

Still, is the ‘accuracy’ of an underlying model necessarily in symmetry with the practical efficiency of

the methods it has spawned? The notion of near-misses and the use of near-miss reporting might

despite the shortcomings of the underlying model prove to be useful in accident prevention work. It

might for example be an effective way to monitor the resonance of normal work and a possibility to

learn more about working conditions and ordinary behavior within a domain or organization.

The accident-ratio models presented in 2.2 have proven to be powerful motivators for near-miss

reporting and proactive work in many domains (Jones et al., 1999). Even though these types of

models probably do not reflect reality but rather simplify it, concepts as that of near-misses do stem

from these models; a concept that has shown to be a helpful aid in reducing accidents in various

domains. The concept of near-misses is therefore a powerful tool that should not be neglected, and

reversely, a concept that should not be taken for granted either. For example, going all the way and

labeling the amount of found near-misses a numerical indicator of industries’ safety awareness

sounds great, but risk being more of a show of statistics than indicative of awareness of something as

broad, complex and rather hard to define as safety.

Accidents and incidents can be understood (or at least feel self-explanatory) without an accident-

ratio model, these concepts still feel like ‘common sense’ to use when assessing the severity of the

outcome or output. The near-miss concept might inherit more of a ‘cause and effect’ point of view,

due to the direct connection to the research of Heinrich et al. (1980) and Bird and Loftus (1976).

Still, the near-miss concept might still be valid and usable from a systemic perspective, it is rather

what kind of factors you ‘collect’ or highlight from an event labeled near-miss; should a near-miss be

regarded as an accident with the last link in the casual chain removed or as an relevant insight of

ordinary situations that you could learn from? A ‘stereotypical’ near-miss might be an event where

the accident is very close at hand, e.g. similar to the event of the train that almost collides with a

vehicle on the tracks described in 2.2. This type of event makes sense in a cause and effect view of

near-misses; where one or a few factors remove the last link in the casual chain and thus prevent the

accident. On the other side of the spectrum are seemingly trivial or diffuse near-misses where this

‘casual chain’ are hard to spot. Does this argue for different categories of near-misses or is the first

55

example with the near-miss on the railroads an illusion of causality (in the sense that the view of a

casual chain of events is too simple an explanation)? To accept the above-mentioned train related

near-miss as causal might cloud the identification of further relevant factors that will not be regarded

as important when analyzing the event through a sequential accident model. Near-misses of a

(seemingly) trivial or diffuse character might feel more natural to regard as resonance in a system

even though both stereotypical and diffuse near-misses perhaps should be analyzed from a systemic

perspective.

At the same time as near-miss reporting makes it possible to conduct quantitative analysis to find

patterns and estimate accident prevention costs, the same potential large quantitative of data could

visualize resonance in a system. The proactive power of near-miss reporting might be as present in a

systemic perspective as in predominant causal perspectives on accident prevention. The usefulness

might then correspond to what factors a report of a near-miss event include and how to efficiently

use this information during analysis. Near-miss reporting in this sense might be a tool to interpret

interconnections (in a broad and not causal sense) between different parts in a system.

6.3 Methodology To gain knowledge of a domain or a specific phenomenon therein through qualitative methods time

is an important factor. This study had limited resources time wise, which made interviews, combined

with a small literature overview to delve deeper into near-misses concept, a suitable choice to find

out more about near-miss reporting. The versatility and the low resource cost of interviewing as

method made it possible to interview both DPs and officers within a relatively short time span.

Meeting people who talk about near-miss reporting is a more time friendly task than actually being

onboard observing people encounter these events in their ordinary work. Furthermore, in the

context of the study's aim, to make barriers that might hinder or prevent near-miss reporting salient,

the choice of semi-structured interviews felt like an appropriate option. The semi-structure of the

interviews made it possible for me to cover certain issues that I deemed relevant but at the same

time giving room for new questions and interesting directions that I did not think of beforehand. The

literary overview gave me some insight of what could be interesting to ask, even though the

possibility to adapt to the respondent was very important due to my limited knowledge of (and

outsider status in) the maritime domain.

Participant observations onboard might have generated more and deeper insights on crew member's

view on reporting given that time and resources were sufficient. This combined with extended

interviews with more persons onboard, rather than only the deck officer, would have been

interesting as well if the study indeed had been a full ethnography, with a far longer time span. Near-

misses are not reported on daily basis, or even weekly, as described by several companies which

limited observation as a suiting choice for this study. Focus groups could have been an interesting

option and dynamic and interactive investigative tool if crew members had participated in the study.

Focus groups could potentially trigger a more thorough discussions regarding near-miss reporting,

though there could be potential problems to have focus groups consisting of DPs from different

companies due to competition and uneasiness to share sensitive information among companies in

the trade. There might also be other reasons that speak against focus groups in certain cases, e.g.

there might be hard to get the necessary participants, or the nature of the questions asked might fit

better in an interview format with a single individual. Beoree (2010) also mentions that strong

56

individuals might affect the focus group and prod others members in directions they would not have

taken otherwise.

The limited amount of time and resources to visit ships that were stationed far from my location

made it necessary to conduct interviews by e-mail in one occasion and interview by telephone in

several occasions. The same restrictions made focus groups a less attractive choice. The e-mail

questions differed from the ordinary interview template in some ways, even though the essence of

the interview was preserved. There were some evident differences in the reply from the e-mailing

officer. The written answers were both shorter and more structured. The e-mail format also limited

any form of further interaction such as follow-up questions on potential relevant threads and made

repairs of misunderstandings impossible. In one occasion a written statement felt a bit ambiguous,

which made the statement hard to interpret given that the intentions and interpretations of the

respondents does not communicate through the media. However, as a mean to discover how near-

misses are viewed by the officers I still deemed that the written account did not differ too much from

spoken ones.

The semi-structure of the interviews - with some questions chosen beforehand that might have

guided the respondents’ train of thought - have most likely affected the answers and also my

expectations of the respondents’ answers in some ways due to the inseparability of methods and

findings (Emerson et al., 1995). In some occasions, interesting notions brought up by the respondents

might also have shaped my follow-up questions, giving them more characteristics of leading

questions than I first intended. It is therefore important to have this in mind when reading the

results, regardless in what form the interviews were conducted, they influence what is seen,

experienced and learned.

Cited source material might change or lose its original meaning when put in different context

(Emerson et al., 1995), as seen in various media coverage and ‘yellow journalism’. The statements I

present in analysis 1 in sections 5.2 and 5.3 has been transcribed, paraphrased and translated. This

makes the transformed material stray from its source, and risks regarding the participants meaning

must be made salient. This kind of transformation might pose a threat to the credibility of the

material, due to the risk that I might influence the material unknowingly when I choose how to

paraphrase and translate the material.

The transferability of this study is difficult to determine. It is hard to label the description of the

material as thick when the gathered data stems from interviews only, and not from other types of

qualitative methods as observations and the like. This even though the aim and research question

easily could be generalized and used within another domain and in other socio technical systems.

Due to the nature of work in complex environments, the concepts used in this study are both taken

from and could be applicable to various domains. The conclusions drawn from this material could be

of gain to other domains in the same way findings in other domains are of use in this study.

Other issues that might come from an interviewing method only, concerns the quixotic reliability

(Lützhöft, 2004), and the risk that some answers are rehearsed, knowingly or unknowingly. Might

the respondents feel uneasy giving their view, presenting issues and seemingly sensitive data to an

interviewer who is not well grounded in the domain? One possible advantage that might limit

rehearsed answers is my neutral status. A neutral status in the sense of that the study is not affiliated

57

with any shipping company in the maritime trade, neither to media nor to other external businesses

that affect the respondents directly.

One way to test the synchronic reliability as mentioned by Lützhöft (2004) would perhaps been to

assign more than one interviewer when interviewing shipping companies or to use more than one

form of interviewing technique on different respondents. It would still be problematic to test for

synchronic reliability when using interviews only and not observations. Two observers can take note

of the same event simultaneously whereas the same thing is not possible when interviewing a

respondent.

I made the question template that was used during the semi-structured interviews as an outsider to

the maritime domain. A risk to consider is that an outsider might not be trusted enough to get the

answers that corresponds the most to the respondents point of view but rather be presented with

‘rehearsed’ or official answers that would strengthen the quixotic reliability and give an misleading

conception of the matter at hand. Still, the reverse could possibly be true as well. A person unfamiliar

with the domain at hand could be given more thoroughly answers to certain questions, due to the

respondent’s belief that a more exhaustive answer is needed.

In retrospect some questions in the question template might have been rephrased, some questions

were latter deemed unnecessary and some threads could have been followed more thoroughly to

gain an even clearer picture and a better understanding of near-miss reporting. The time lapse

between the conducted interviews led to a knowledge gain for me as an interviewer during the

course of interviewing all respondents. Some threads, that I found relevant during this process, were

explored more in the latter interviews, due to my lack of domain knowledge beforehand. However,

there are not necessarily only drawbacks having the outsider status, due to potential insider bias

some questions in the template (see Appendix A), might have been viewed as self-explanatory for

insiders and might have been left out had I known more about the domain beforehand.

Analysis 1 was made in a naturalistic fashion where found themes are connected to the answers

given by the respondents. It is hard to know whether the themes presented in this study are the

most representative ones for barriers to near-miss reporting in the maritime domain. The themes in

themselves, they are affected by the structure of my interview template and by my limited

knowledge in the domain; in a sense of what I ask is what I get. Another question template could

possibly create other themes. In the same sense are the generalization and categorization of

statements into Barach and Small’s framework in analysis 2 my own choice. Many of Barach and

Small’s (2000) subcategories were not found in this material and one main category, the societal

category, was not used at all as the collected data was not deemed broad enough to enable an

analysis from the societal point of view. This does not necessarily indicate that these categories are

unimportant but rather that the question templates used for interviewing were not tailored to make

visible barriers and incentives in all given areas. None of Barach and Small’s categories were defined

in their study which made it difficult to know the authors intent in regard to some of the categories,

this led to my own interpretation of these categories instead.

There is also a question of whether the material gathered is representative of the view of the

respondents. The use of interviews is fast and effective but does not give more information than the

respondents knowingly or unknowingly choose to share with the interviewer. This create a situation

58

where the point of view constructed by me is judged against the view of the respondent, do I

correctly present a sound description of the respondents mentioned views on reporting? Perceived

and highlighted barriers in the material are not only those that are mentioned directly by the

respondents, some of the perceived barriers stem from my own interpretation of the respondents

answers. Without the use of other methods and triangulation, it is hard to reach the next layer; even

if I do a sound and well described analysis of statements and overarching themes, this presentation

has not been juxtaposed against findings derived from other methods and it is difficult to know

whether the respondents are presenting the ‘correct view’. The respondents, knowingly or

unknowingly, might have left out parts of their views, perhaps integral parts of their worldview that

cannot be made salient with interviews only. The interviews do although show differences in views

between DPs and officers and similarities within the respondent groups, this might be seen as an

indicator of objectiveness in the sense of potentially reliable insights in their point of view.

Do I, as an outsider, view, analyze and present the gathered material as would a person with more

domain knowledge and professional vision do? One way to investigate this could be to let a insider

re-analyze the raw material, but It could at the same time exist individual and cultural differences

within the domain that would make it hard to draw conclusions regarding my ‘correctness’ from such

an investigation. If my own analysis differ from the analysis of the insider, this would not necessarily

imply that my view is ‘wrong’.

6.4 Further research The parallels between the maritime domain and other socio technical systems found in this study

could be further explored, perhaps by more extensive use of qualitative methods such as

observations or a full ethnography. Cross-domain studies are of course of importance to detect both

similarities and potential domain specific factors. Maintaining safety in aviation or in the maritime

domain might differ from, for example, patient safety in the medical domain. The process of safely

transporting a vehicle or vessel from one point to another differs from treating people medically,

whereas a more fitting resemblance might be mechanical work. Is it possible to discern differences in

different domains looking at them functionally?

The discussion of the first analysis suggests that analyzing of the reporting process in INSJÖ could be

of relevance, both to assess the design of this reporting system, and the means to report. It would

further be of relevance to explore what type of accident etiology that is the occupational norm in the

maritime domain. If a storytelling scheme, as described by Sanne (2007), were present in this domain

would it be possible to harvest the positive sides of this norm and incorporate those aspects in a

broader reporting scheme?

The present study does not focus on the view of the rest of the crew beside the reporting officer

onboard; it is unclear whether the crew members onboard feel that they get adequate feedback

from officers and the DP. Some officers mention that crew members of certain nationalities report

less or not at all or that differences in crew members’ background might cloud the larger safety

picture. These threads might be interesting to follow up if conducting interviews with crew members

in future studies.

This report does not scrutinize the term safety culture, even though several participants mention the

importance of good safety culture and how a no-blame view is prevalent in one's company. How

59

important a brick near-miss reporting is in achieving a sound safety culture remains unanswered in

this study.

60

7 Conclusions

The last chapter presents the conclusions of the study, with the aim of identifying and examining

near-miss reporting and its potential barriers within the Swedish maritime domain. The following

highlighted conclusions are derived from the analyses and discussions of the collected data.

There are tendencies of different philanthropic values among shipping companies. Some

companies value information sharing beneficial to the maritime domain as a whole, in contrast

to sharing information only locally within their own company or ships with similar setup. This

might create repercussions regarding the efficiency of near-miss reporting as proactive accident

prevention.

The therapeutic factor, to teach and learn from others is held high by a majority of the

respondents.

The perceived views of near-miss reporting in this study and the potential differences are

between DPs and officers. The other crew members’ view remains unknown.

To make reporting effective, and decrease the risk of unbalance between rehearsed benefits of

reporting and reality, it is important to give personnel ownership of their own reporting system

and the knowledge how and why to use it.

An interesting view of anonymity and no blame culture is made salient. It is deemed important

to provide anonymity but the need among the crew should not be strong if indeed a no blame

culture and a sound safety culture exist in the company.

Do external influences, such as Oil companies, add a potential risk of urging the company in

finding debatable near-misses out of necessity due to strict requirements?

To present reports in manner that is deemed useful for people in the maritime trade might be

important to minimize the risk of crippling proactive safety work.

Further studies of the reporting process and if there is a difference in accident etiologies, such as

storytelling and a more organizational incident scheme could be of importance, to learn more

regarding reporting in the maritime domain.

The accident-ratio models depicted by Bird and Loftus (1976) and Heinrich et al. (1980) might be

too simplistic models to explain why accidents occur, but might nevertheless be used as

powerful tools and incentives for near-miss reporting.

61

References Barach, P. & Small, S.D. (2000). Reporting and preventing medical mishaps: lessons from non-medical

near miss reporting systems. BMJ, 320, 759-763.

Benyon, D. ,Turner, P., Turner, S. (2005). Designing Interactive Systems. People, Activities, Contexts,

Technologies. Essex: Pearson Education Limited.

Bird, F.E., Loftus, R.G. (1976). Loss Control Management. Loganville: Institute Press, Division of

International Loss Control.

Bryman, A. (2002). Samhällsvetenskapliga metoder. Malmö: Liber Ekonomi.

Cambraia, F.B., Saurin, T.A., Formoso, C.T. (2010). Identification, analysis and dissemination of

information on near misses: A case study in the construction industry. Safety Science, 48(2010), 91-

99.

Dekker, S. (2010). In the system view of human factors, who is accountable for failure and success? In

D. de Waard, A. Axelsson, M. Berglund, B. Peters, and C. Weikert (Eds.), Human Factors: A system

view of human, technology and organisation (pp09-27). Maastricht: Shaker Publishing.

Elder, N.C., Graham, D., Brandth, E. & Hickner, J. (2007). Barriers for Making Error Reports from

Family Medicine Offices: A Report from the American Academy of Family Physicians National

Research Network. The Journal of the American Board of Family Medicine, 20, 115-123.

Emerson, R. M., Fretz, R.I., Shaw, L.L. (1995). Writing Ethnographic Fieldnotes. Chicago: The

University of Chicago Press.

Evans, S. M., Berry, J.G., Smith, B. J., Esterman, A., Selim, P., O’Shaughnessy, J. & DeWit, M. (2006).

Attitudes and barriers to incident reporting: a collaborative hospital study. Quality and Safety in

Health Care, 15, 39-43.

Geertz, C. (1973). The interpretation of cultures. New York: Basic books.

Goodwin, C. (1994). Professional vision. American Anthropologist, 96(3), 603-633.

Heinrich, H.W., Petersen, D., Roos, N. (1980). Industrial Accident Prevention – Safety Management

Approach, 5th ed. London: McGraw-Hill, Inc.

Heiman, G.W. (2001). Understanding Research Methods and Statistics – An Integrated Introduction

for Psychology, 2nd ed. Boston and New York: Houghton Mifflin Company.

IMO. (2007). Guidance on the qualifications, training and experience necessary for undertaking the

role of the designated person under the provisions of the International Safety Management (ISM)

code. The International Maritime Organization.

IMO. (2008). Guidance on Near-Miss Reporting. The International Maritime Organization.

Johnson, C.W. (2003). Failure in Safety-Critical Systems: A Handbook of Accident and Incident

Reporting. Glasgow: University of Glasgow Press.

62

Jones, S., Kirchsteiger, C. & Bjerke, B. (1999). The importance of near miss reporting to further

improve safety performance. Journal of Loss Prevention in the Process Industries, 12, 59-67.

Jordan, P.W. (1998). An introduction to Usability. London: Taylor & Francis.

Knoublauch, H. (2005). Focused Ethnography. Forum: Qualitative Social Research, vol 6, No. 3. Art 44.

Lützhöft, M., Nyce, J. M., Styhr Petersen, E. (2009). Epistemology in ethnography: assessing the

quality of knowledge in human factors research. Theoretical Issues in Ergonomics Science, 1-14.

Lützhöft. M. (2004). “The technology is great when it works” – Maritime Technology and Human

Integration on the Ship’s Bridge. Doctoral thesis, University of Linköping, Linköping.

MAIB. (1987). Herald of Free Enterprise report. Marine Accident Investigation Branch.

Perrow, C. (1999). Normal Accidents – Living with High-Risk Technologies. New Jersey: Princeton

University Press.

Reason, J. (1990). Human Error. New York: Cambridge University Press.

Sanne, J.M. (2007). Incident reporting or storytelling? Competing schemes in a safety-critical and

hazardous work setting. Safety Science, 46, 1205-1222.

van der Schaaf, T. & Kanse, L. (2003). Biases in incident reporting databases: an empirical study in the

chemical process industry. Safety Science, 42, 57-67.

Zachau, J. (2008). Near-Misses and Accidents in Proactive Safety Work. A study of human and other

factors in near-miss and accident databases. Retrieved from http://www.transportstyrelsen.se

Internet Sources.

IMO. (2002). About IMO. Retrieved 2010-04-08 from

http://www.imo.org/

Boeree, G.C. (2010). Qualitative Methods Workbook. Retrieved 2010-02-22 from

http://webspace.ship.edu/cgboer/qualmeth.html

CHIRP. (2007). About CHIRP. Retrieved 2010-05-24 from

http://chirp.co.uk/information-about.asp

EMSA. (2010). The European Maritime Safety Agency: Its origin and its tasks. Retrieved 2010-03-08

from

http://www.emsa.europa.eu/end179d002.html

IMO. (2002). The ISM Code. Retrieved 2009-11-20 from

http://www.imo.org/humanelement/mainframe.asp?topic_id=287

IMO. (2002). Prevention of pollution by oil. Retrieved 2010-06-16 from

http://www.imo.org/environment/mainframe.asp?topic_id=231#9

63

INSJÖ. (2005). About INSJÖ. Retrieved 2010-05-24 from

http://insjo.org/Om_INSJOE/Om_INSJOE.asp

INSJÖ. (2007). Vilka organisationer står bakom INSJÖ? Retrieved 2010-05-24 from

http://insjo.org/FAQ/FAQVilkaorganisationerstrbakomINSJ.asp

INSJÖ. (2010). Erfarenhetsbanken. Retrieved 2010-05-24 from

http://insjo.org/Erfarenhetsbanken/Erfarenhetsbanken.asp

Purdue Online Writing Lab. (2010). Quoting, Pharaphrasing and Summarizing. Retrieved 2010-03-08

from

http://owl.english.purdue.edu/owl/resource/563/1/

IMO. (2002). Safety of ro-ro ferries. Retrieved 2010-04-08 from

http://www.imo.org/Safety/mainframe.asp?topic_id=1536

MARS. (2008). About MARS. Retrieved 2010-05-24 from

http://www.nautinst.org/mars/about.htm

Sjöfartsverket. (2010). About us. Retrieved 2010-05-24 from

http://www.sjofartsverket.se/en/About-us/

Sveriges Redareförening. (2010). Sjösäkerhet. Retrieved 2010-05-24 from

http://www.sweship.se/Sveriges_Redareforening_Sjosakerhet___Teknik_Sjosakerhet_DXNI-

24641_.aspx

Transportstyrelsen. (2008). Statistiksammanfattning. Retrieved 2010-05-24 from

http://transportstyrelsen.se/sv/Sjofart/Olyckor--tillbud/Statistiksammanstallning/

Transportstyrelsen. (2009). Accidents & Near-Misses. Retrieved 2010-05-24 from

http://www.transportstyrelsen.se/en/Shipping/Accidents--Near-Misses/

Transportstyrelsen. (2010). Sjöfartsavdelningen informerar ... 2/2010. Retrieved 2010-06-16 from

http://www.transportstyrelsen.se/sv/Sjofart/Olyckor--tillbud/Sjofartsinspektionen-informerar/

64

Appendix A - Interview template

Frågor till rederier som rapporterar till INSJÖ

Frågor till DP:

Synen på near-misses

Hur ofta eller i vilken utsträckning brukar fartygen rapportera in händelser som kunde ha lett till en incident eller olycka? Är denna typ av händelser eller tillbud något du känner igen från säkerhetsarbetet på rederiet?

Är denna typ av rapportering något du tycker ska utökas eller förbättras? - Hur?

Hur viktig anser du att denna typ av rapporter är? - Varför?

Synen på och syftet med incidentrapportering

Beskriv hur säkerhetsarbetet ser ut för ert rederi.

Hur stor vikt ligger på rapportering av olyckor, incidenter och tillbud i säkerhetsarbetet?

Varför är det viktigt att rapportera olyckor, incidenter och tillbud?

Hur viktig är denna form av rapportering i ditt arbete som DP? - Varför?

Får du någon form av feedback på inrapporterade händelser, till exempelvis genom INSJÖ? - Hur använder du i så fall denna feedback i ditt säkerhetsarbete?

Hur viktig anser du att anonymitet och möjligheten till att vara konfidentiell vid inrapportering är? - Varför?

Är dessa faktorer något som påverkar detta din inrapportering (till INSJÖ) på något sätt? - Hur?

Synen på DP:s roll

Beskriv hur säkerhetsarbetet ser ut för dig som DP.

Vilka är dina viktigaste uppgifter gällande säkerheten?

Hur sker kommunikationen med rederiets fartyg?

Hur skulle kommunikationen kunna förbättras med rederiets olika fartyg?

Trust mellan DP och besättning?

Känner du att det råder ett öppet klimat inom rederiet gällande säkerhetsarbetet?

Känner du att besättningen rederiets fartyg förstår din roll som DP? - Varför?

Känner du att besättningen rederiets fartyg värderar din roll som DP?

65

- Varför?

Beskriv kommunikationen mellan fartygen och dig som DP. Vad för typ av information förmedlas oftast till dig från fartygen?

Vad beror eventuellt motstånd till rapportering på?

Känner du dig nöjd med mängden inrapporterade händelser från fartygen?

Vad tror du detta beror på?

Hur skulle man kunna öka inrapporterandet av händelser från fartygen tror du?

Vilka faktorer eller barriärer tror du kan hindra eller påverka antalet inrapporterade händelser från fartygen? - Varför?

Finns det händelser inrapporterade från fartygen som du anser är onödiga att föra vidare in till INSJÖ?

- Varför? - Vilka typer av händelser rör det sig om?

Synen på och möjligheten till Feedback på rapportering.

Hur amvänds den feedback som fås från INSJÖ?

Tycker du att det är något som saknas i den feedback du får från INSJÖ? - I så fall vad?

66

Frågor till rederier som rapporterar till INSJÖ. Frågor till besättning:

Synen på near-misses

Vad anser du vara en olycka?

Vad anser du vara en incident?

Vad anser du vara ett tillbud?

Hur ofta eller i vilken utsträckning brukar ni rapportera händelser som kunde ha lett till en incident eller olycka?

Hur viktig anser du att det är att rapportera denna typ av rapporter? - Varför?

Synen på och syftet med incidentrapportering

Beskriv hur säkerhetsarbetet ser ut för ert rederi.

Hur stor vikt ligger på rapportering av olyckor, incidenter och tillbud i säkerhetsarbetet?

Varför är det viktigt att rapportera olyckor, incidenter och tillbud?

Hur viktig anser du att anonymitet och möjligheten till att vara konfidentiell vid inrapportering är? - Varför?

Är dessa faktorer något som påverkar detta din inrapportering på något sätt. - Hur?

Synen på DP:s roll

Beskriv säkerhetsarbetet ombord på fartyget.

Genom vilka kanaler sker säkerhetsarbetet ombord?

Vilken är DP:s funktion inom rederiet?

Hur viktig är DP:s roll för fartygets säkerhet?

Vilken form av feedback får du på inrapporterade händelser och hur används denna feedback?

Beskriv kommunikationen med DP.

Hur skulle kommunikationen kunna förbättras med DP?

Trust mellan DP och besättning?

Känner du att det råder ett öppet klimat inom rederiet gällande säkerhetsarbetet?

Vem vänder du dig till om du upptäcker brister i säkerheten?

Synen på ansvar (Loss of face?)

Finns det händelser eller tillfällen när du tvekar att rapportera vissa händelser ombord?

Vad skulle detta kunna bero detta på?

Vad kan konsekvenserna (i värsta fall) bli att rapportera känsliga händelser?

Vad kan konsekvenserna (i värsta fall) bli att inte rapportera känsliga händelser?

67

Synen på och möjligheten till Feedback på rapportering (både från besättningens och DP:s sida).

Hur viktig anser du att anonymiteten och konfidentialiteten vid inrapportering är?

Är dessa faktorer något som påverkar detta din inrapportering på något sätt. - Hur?

Känner du dig delaktig i det övergripande säkerhetsarbetet inom rederiet?

Ger du själv – eller har du möjlighet att ge - förslag på förbättringar till rederiet / DP?

Känner du att du får tillräcklig feedback från DP på inrapporterade händelser och annat som rör säkerhetsarbetet inom rederiet?

68

Frågor till rederier som inte rapporterar till INSJÖ Frågor till DP:

Synen på near-misses

Vad anser du vara en olycka?

Vad anser du vara en incident?

Vad anser du vara ett tillbud?

Hur ofta eller i vilken utsträckning brukar fartygen rapportera in händelser som kunde ha lett till en incident eller olycka? Är denna typ av händelser eller tillbud något du känner igen från säkerhetsarbetet på rederiet?

Är denna typ av rapportering något du tycker ska utökas eller förbättras? - Hur?

Hur viktig anser du att denna typ av rapporter är? - Varför?

Synen på och syftet med incidentrapportering

Beskriv hur säkerhetsarbetet ser ut för ert rederi.

Hur stor vikt ligger på rapportering av olyckor, incidenter och tillbud i säkerhetsarbetet?

Varför är det viktigt att rapportera olyckor, incidenter och tillbud?

Hur viktig är denna form av rapportering i ditt arbete som DP? - Varför?

Hur används de rapporter som kommer in från fartygen?

Skickas vissa rapporter vidare till en annan part?

Hur viktig anser du att anonymitet och möjligheten till att vara konfidentiell vid inrapportering är? - Varför?

Synen på DP:s roll

Beskriv hur säkerhetsarbetet ser ut för dig som DP.

Vilka är dina viktigaste uppgifter gällande säkerheten?

Hur sker kommunikationen med rederiets fartyg?

Hur skulle kommunikationen kunna förbättras med rederiets olika fartyg?

Trust mellan DP och besättning?

Känner du att det råder ett öppet klimat inom rederiet gällande säkerhetsarbetet?

Känner du att besättningen rederiets fartyg förstår din roll som DP? - Varför?

Känner du att besättningen rederiets fartyg värderar din roll som DP? - Varför?

Beskriv kommunikationen mellan fartygen och dig som DP. Vad för typ av information förmedlas oftast till dig från fartygen?

69

Vad beror motvilja till rapportering på?

Känner du dig nöjd med mängden inrapporterade händelser från fartygen?

Vad tror du detta beror på?

Hur skulle man kunna öka inrapporterandet av händelser från fartygen tror du?

Vilka faktorer eller barriärer tror du kan hindra eller påverka antalet inrapporterade händelser från fartygen? - Varför?

Synen på och möjligheten till Feedback på rapportering.

Vad för slags feedback får du på ditt säkerhetsarbete?

Är denna feedback tillräcklig?

70

Frågor till rederier som inte rapporterar till INSJÖ

Frågor till Besättning:

Synen på near-misses

Hur ofta eller i vilken utsträckning brukar ni rapportera händelser som kunde ha lett till en incident eller olycka?

Hur viktig anser du att det är att rapportera denna typ av rapporter? - Varför?

Synen på och syftet med incidentrapportering

Beskriv hur säkerhetsarbetet ser ut för ert rederi.

Hur stor vikt ligger på rapportering av olyckor, incidenter och tillbud i säkerhetsarbetet?

Varför är det viktigt att rapportera olyckor, incidenter och tillbud?

Hur viktig anser du att anonymitet och möjligheten till att vara konfidentiell vid inrapportering är? - Varför?

Är dessa faktorer något som påverkar detta din inrapportering på något sätt. - Hur?

Synen på DP:s roll

Beskriv säkerhetsarbetet ombord på fartyget.

Genom vilka kanaler sker säkerhetsarbetet ombord?

Vilken är DP:s funktion inom rederiet?

Hur viktig är DP:s roll för fartygets säkerhet?

Vilken form av feedback får du på inrapporterade händelser och hur används denna feedback?

Beskriv kommunikationen med DP.

Hur skulle kommunikationen kunna förbättras med DP?

Trust mellan DP och besättning?

Känner du att det råder ett öppet klimat inom rederiet gällande säkerhetsarbetet?

Vem vänder du dig till om du upptäcker brister i säkerheten?

Synen på ansvar (Loss of face?)

Finns det händelser eller tillfällen när du tvekar att rapportera vissa händelser ombord?

Vad skulle detta kunna bero detta på?

Vad kan konsekvenserna (i värsta fall) bli att rapportera känsliga händelser?

Vad kan konsekvenserna (i värsta fall) bli att inte rapportera känsliga händelser?

Synen på och möjligheten till Feedback på rapportering (både från besättningens och DP:s sida).

71

Hur viktig anser du att anonymiteten och konfidentialiteten vid inrapportering är?

Är dessa faktorer något som påverkar detta din inrapportering på något sätt. - Hur?

Känner du dig delaktig i det övergripande säkerhetsarbetet inom rederiet?

Ger du själv – eller har du möjlighet att ge - förslag på förbättringar till rederiet / DP?

Känner du att du får tillräcklig feedback från DP på inrapporterade händelser och annat som rör säkerhetsarbetet inom rederiet?